Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Are young children able to learn exploratory strategies by observation?

Are young children able to learn exploratory strategies by observation? Psychological Research (2018) 82:1212–1223 https://doi.org/10.1007/s00426-017-0896-0 O R I G IN AL ARTI CL E Are young children able to learn exploratory strategies by observation? 1,2 3 3,4 3 • • • • Francesca Foti Domenico Martone Stefania Orru` Simone Montuori 4 3,4 2,5 2,3,6 • • • Esther Imperlini Pasqualina Buono Laura Petrosini Laura Mandolesi Received: 21 December 2016 / Accepted: 14 July 2017 / Published online: 20 July 2017 The Author(s) 2017. This article is an open access publication Abstract New competencies may be learned through children who directly performed the RAM task without any active experience (experiential learning or learning by observation. The main result of the present research is that doing) or observation of others’ experiences (learning by the children who observed the highly structured and correct observation). Observing another person performing a exploratory strategy spent less time, made fewer errors, complex action facilitates the observer’s acquisition of the exhibited a longer spatial span, and thus they explored the same action. The present research is aimed at analyzing if maze more efficiently than the children who directly per- the observation of specific explorative strategies adopted in formed the RAM task without any observation. This finding a constrained environment, such as the Radial Arm Maze indicates that when the observed explorative procedure is (RAM), could help young children to explore the maze and structured, sequential and repetitive the action understand- to build a cognitive spatial map of the explored environ- ing and information storage processes are more effective. ment. To this aim young children were randomly assigned Importantly, the observation of specific spatial strategies to three groups: children who performed the RAM task helped the children to build the cognitive spatial map of the following the observation of an actor solving the same maze explored environment and consequently to acquire/enrich by putting into action a highly structured exploratory the declarative knowledge of the environment. strategy; children who performed the RAM task following the observation of the actor solving the same maze by putting into action a less structured exploratory strategy; Introduction New competencies may be learned through active experi- Francesca Foti and Domenico Martone contributed equally to this work. ence (experiential learning or learning by doing) or observation of others’ experiences (learning by observa- & Laura Mandolesi tion) (Bandura, 1977; Meltzoff, Kuhl, Movellan, & Sej- laura.mandolesi@uniparthenope.it nowski, 2009). Department of Medical and Surgical Sciences, University Learning by observation does not just involve copying ‘‘Magna Graecia’’, Catanzaro, Italy an action, but it requires that the observer transforms the IRCCS Fondazione Santa Lucia, Via del Fosso di Fiorano 65, observation into an action as similar as possible to the 00143 Rome, Italy model in terms of the goal to be reached and motor Department of Movement Sciences and Wellbeing, strategies to be applied (Meltzoff & Andrew, 1995; Melt- University ‘‘Parthenope’’, Naples, Italy zoff & Decety, 2003). Observing another person perform- ing a complex action represents a desirable condition of Fondazione IRCCS SDN, Naples, Italy learning that enables the learner to better understand the Department of Psychology, ‘‘Sapienza’’ University of Rome, skill prior to the performance and/or it helps the learner to Rome, Italy 6 more readily discriminate perceptually variables that are Department of Motor Science and Wellbeing, University important for the performance of that skill (Bird & Heyes, ‘‘Parthenope’’, Via Medina, 40, 80133 Naples, Italy 123 Psychological Research (2018) 82:1212–1223 1213 2005; Meltzoff et al., 2009). It is believed that observation solution. The findings of these researches suggest that of an action facilitates motor learning of that skill because young children are selective copiers who reproduce the it facilitates the acquisition of the main spatial and tem- irrelevant tool actions most frequently after having viewed poral features of the task, and thus removes the need to high-status models performing them. Thus, learning by create a cognitive representation of the action pattern observation represents a learning mechanism that can be through experiential learning (Keetch, Schmidt, Lee, & used in several fields (e.g., school and sport) as a ‘‘learning Young, 2005; Buchanan & Dean, 2010; Rohbanfard & technique’’. In addition, several studies have highlighted Proteau, 2011). However, it is worth of noting that condi- the importance of the observational learning in children tions of learning that accelerate the learning, by limiting with intellectual disabilities (Foti et al., 2013, 2014, 2015). the time-consuming process of learning by trial and error As for its neurobiological basis, the learning by obser- and reducing the practice needed to learn, often fail to vation is thought to utilize brain regions responsive to both support long-term retention and transfer (Schmidt & Bjork, observation and execution of action, as the mirror neuron 1992; Bjork, 2011; Bjork, Dunlosky, & Kornell, 2013). system (Gallese, Fadiga, Fogassi, & Rizzolatti, 1996; Acquiring skills by observation is a fundamental cog- Rizzolatti, Fogassi, & Gallese, 2001; Rizzolatti & nitive ability already existing from the birth (Meltzoff & Craighero, 2004). The mirror neuron system includes pre- Moore, 1977; Nadel & Butterworth, 1998; Meltzoff et al., motor cortex, inferior frontal gyrus, and inferior parietal 2009; Nadel, 2002). Already at 18-months-old children lobule, areas which receive their main visual input from the may learn a novel motor pattern by observation (Herold & superior temporal sulcus (Molenberghs, Brander, Mattin- Akhtar, 2008; Matheson, Moore, & Akhtar, 2013) and if gley, & Cunnington, 2010; Caspers, Zilles, Laird, & the adults explicitly show their intention prior to demon- Eickhoff, 2010). Insofar as it generates a simulation circuit stration, even 16-months-old infants learn by observation that allows the association between one’s own actions with (Fagard, Rat-Fischer, Esseily, Somogyi, & O’Regan, others’ actions, the mirror neuron system is retained to be 2016). Three-year-old children are able to learn how to involved in action understanding, imagination, and imita- extract a reward from a box following a video-demon- tion (Rizzolatti & Sinigaglia, 2010), and thus even in the stration of the correct procedure (Flynn & Whiten, 2013). observational learning. Besides imitative abilities learning by observation Most developmental studies focused on how and what requires cognitive competencies, as attentive and mnesic the observer child has to observe to promote learning functions, sequencing abilities, planning, response inhibi- (Rohbanfard & Proteau, 2011; Marshall & Meltzoff, 2014; tion, cognitive flexibility, good knowledge and anticipatory Carr et al., 2015). However, to our knowledge there are no expectation of effects related to actions, goal-directed developmental studies that investigated whether the actions, and motor imagery allowing recombination of learning by observation of exploratory strategies promotes novel actions with novel effects (Foti et al., the acquisition of navigational abilities. For this reason, we 2013, 2014, 2015; Torriero, Oliveri, Koch, Caltagirone, & wondered if observing an adult actor who adopts specific Petrosini, 2007). Furthermore, to learn by observation it is navigational strategies to explore a radial arm maze (RAM) necessary to observe and attend to the actor, engage in joint can help young children improve their exploration of the attention, understand and reproduce other’s actions. Thus, same maze and build the cognitive spatial map of the learning by observation also represents a powerful social explored environment. Another aim of the present research learning mechanism (Frith & Frith, 2012). For example, is to determine whether observation of a structured model children can learn how to behave in social contexts by or of a less structured model of explorative strategies observing how adults interact with each other (Shimpi, would have resulted in different reproduction of the Akhtar, & Moore, 2013). Recently, it was shown that if the explorative patterns. On one hand, it has been proposed model has a high social status, such as a teacher, the that the observation permits the observer to develop a sort children tend to learn even irrelevant information by of ‘‘perceptual blueprint’’ of the task to be learned (Ban- observation (McGuigan, Gladstone, & Cook, 2012; dura, 1977). This may work in favor of the utilization of a McGuigan, 2013) or attempts without outcome (Carr, model performing a very efficient and successful explo- Kendal, & Flynn, 2015). The typical scenario in these rative strategy. On the other hand, it might be fruitful also studies is that before being allowed to attempt the task to observe a model performing a less structured and with a themselves, the observers watch an adult model perform a superior mnesic load, but still successful explorative sequence of tool actions varying according to their causal strategy. To these aims, young children (mean age: 5 years necessity, with some of the actions being necessary for and 3 months) were randomly assigned to three groups: reward retrieval, others being causally irrelevant (as per- children who performed the RAM task following the forming unnecessary taps before retrieving a reward from a observation of an actor solving the same maze by putting box) and others without the efficacy of an observed into action a highly structured and efficient exploratory 123 1214 Psychological Research (2018) 82:1212–1223 strategy (such as sequentially entering adjacent arms); children who performed the RAM task following the observation of the actor solving the same maze by putting into action a less structured and still successful exploratory strategy (such as randomly entering arms); children who directly performed the RAM task without any observa- tional training. Methods Participants Thirty-six healthy Italian children (17 M and 19 F) aged from 4 years and 6 months (4.6) to 5 years and 9 months (5.9) (mean age ± SD 5.3 ± 0.2) participated in the pre- sent study. Children were subdivided into three groups according to the following experimental conditions: Learning by Observation of a highly Structured explorative Fig. 1 Schematic representation of the agility course strategy (LeOS) (N = 11; 5 M and 6 F; mean age 5.2 ± 0.3); Learning by Observation of Random explo- rative strategy (LeOR) (N = 13; 6 M and 7 F; mean age Slaloming: the child runs in a zig zag pattern among six cones; Crawling: the child grovels under a rod held by two 5.3 ± 0.2); Learning by Doing (LeD) (N = 12; 6 M and 6 F; mean age 5.3 ± 0.1). No children have had previous cones set at 50 cm (19.68 in) from the ground; Catching: the child enters a circle placed on the ground in which he/ experience with the RAM task. All children had normal or corrected-to-normal vision she grasps a ball thrown by the teacher positioned in front of him/her; Shooting for goal: the child throws the ball into and standard anthropometric measurements and presented the basket located in front of him/her. no neurological or neuropsychological problems. Body Total score (the sum of scores ranging from 0 to 7) and mass index values (M = 17.06 ± 2.8; F = 16.83 ± 1.62) total time (time to perform the entire course) were were between the 50th–75th percentile. To exclude the recorded. presence of sensory-motor deficits, the psychomotor development of all children was evaluated through a bat- tery of exercises of motor accuracy (Niederer et al., 2011). Drawing test of the human figure To verify graphic abilities and cognitive development, all children were assessed in the drawing test of the human According to Machover’s instructions (Machover, 1949), each child was asked to ‘‘draw someone’’. For child’s figure (Machover, 1949). All children attended a kinder- question on what it was possible to draw, the experimenter garten school in South Italy where a 1 h/day of physical replied ‘‘whatever you want’’. If the child drew only the activity was planned for 5 days/week. The parents of head, the investigator encouraged him/her to draw the children gave informed written consent. The study was whole figure. Since the children were less than 6 years of conducted according to the 1964 Declaration of Helsinki. age, the qualitative assessment of drawing human fig- Motor accuracy assessment ure was focused to highlight whether the child did not draw significant details, such as hands, hair, eyes, mouth (Di In the school gym an ‘‘agility course’’ was built (Fig. 1) Leo, 1970; Cox, 1992; Boncori, 2006). where the following seven motor abilities were evaluated assigning ‘‘1’’ or ‘‘0’’ scores according to correctness or Apparatus incorrectness, respectively. The RAM adapted for children consisted of a round central Somersaulting: the child rolls forward in a complete platform [1 m (39.37 in) in diameter] with eight arms revolution around the horizontal axis on a carpet at the start [50 cm (19.68 in) wide 9 3.5 m (137.79 in) long] radiating of the course; Balancing: the child walks heel to toe on a like the spokes of a wheel (Fig. 2). To force the child to white 2 m (78.74 in) tape (10 cm (3.93 in) large) fixed on exit from an arm and return to the center of the starting to the ground; Jumping: the child hops three 25 cm (9.84 in) high obstacles, built with two cones joined by a rod; platform before entering another arm, the sides of each arm 123 Psychological Research (2018) 82:1212–1223 1215 bucket only once. An error was made when the child re- entered an arm already visited during the same trial. Each child performed three trials a day for three consecutive days. Since the three daily trials constituted a session and each child made three sessions, each child performed nine trials. At the end of each trial, the child waited 1 h (inter- trial interval), before being re-tested in the RAM. At the beginning of RAM testing, the experimenter used the same simple verbal instructions to explain the task to each child (‘‘The game is to find the little colored balls. Do you see the colored buckets at the end of each alley? You have to reach a bucket, take the little ball inside, and then go back to the center, where the platform is, until you have col- Fig. 2 View of the eight-arm radial maze lected all the balls. Be careful to reach the buckets always were marked off by white and red ribbons hung across the staying inside the maze. Go and have fun!’’). No other opening and the end of the arm, forming a sort of con- instructions or verbal encouragement were provided during straining barrier. This procedure prevented the children testing. In the two observation conditions (LeOS and from ‘‘cutting corners’’ as they exited from an arm and LeOR), before starting RAM exploration the experimenter forced them to exit and return to the center of the starting told the children: ‘‘The game is to find the little colored platform before entering another arm. At the end of each balls inside the buckets. Look at me carefully’’. In the arm, there was a red plastic bucket (18 cm (7.08 in) LeOS condition, each child observed three sessions of three wide 9 28 cm (11.02 in) high) containing the reward (a trials each in which the actor explored the RAM entering little colored ball). The RAM, located outdoors in a foot- always the adjacent arms and stopped after the eight ball field, was surrounded by extra-maze cues (trees, rewards were collected. In the LeOR condition, each child swings, benches, etc.) held in constant spatial relations observed three sessions of three trials each in which the throughout the experiment. The arms were virtually num- actor explored the RAM using a pseudorandom explorative bered in a clockwise direction, considering arm 1 as the strategy and stopping after eight rewards collected. In farthest from the experimenter’s location. Only during the LeOS and LeOR conditions, children observed the actor at experiment could the children see the maze or have phys- distance of about 1.5 m (59.05 in) from the RAM, chang- ical access to it. To increase the motivation of picking up ing their point of observation at every session. Then, each the rewards, at the end of each trial the child received a child actively experienced the three RAM sessions (RAM reward (a little toy) in exchange for all the colored balls testing; Table 1). The trials were annulled if the child left found in the buckets. the maze. However, very few children of LeD condition engaged in this behavior and, in any case, only in the very Experimental procedure first trials of the task. In LeOS and LeOR conditions, no child left the maze. The RAM testing lasted 3 consecutive The three experimental conditions were:—Learning by days and in this execution phase all children were video- Observation of a highly Structured explorative strategy taped and recorded manually. At the end of RAM testing (LeOS), in which the children performed the RAM task phase, all children were asked to make a drawing of the following the observation of an actor solving the maze setting where they had just ‘‘played’’ to evaluate their through a highly structured exploratory strategy;—Learn- mental representative mapping abilities. ing by Observation of Random explorative strategy (LeOR), in which the children performed the RAM task Behavioral parameters following the observation of the actor solving the maze through a less structured exploratory strategy;—Learning We evaluated:—total time (in seconds) spent to complete by Doing (LeD) in which the children directly performed the task;—entries, calculated as the number of visited the RAM task without observation (Table 1). arms;—errors, calculated as the number of re-entries into In each trial of the RAM testing, each child was allowed already visited arms;—spatial span, calculated as the to explore freely the eight arms to retrieve the reward. A longest sequence of correctly visited arms;—persevera- trial ended when all eight rewards had been collected, 20 tions, calculated as the percentage of consecutive entries choices had been made, or 10 min had elapsed from the into the same arm or the re-entries into a fixed sequence of start of the task. Since the buckets were never rewarded arms, divided by the number of arms visited;—percentage twice, the optimal performance consisted of visiting each of angled turns, calculated as the number of a given angle 123 1216 Psychological Research (2018) 82:1212–1223 Table 1 Experimental procedures of the three experimental conditions Day 1 Day 2 Day 3 Day 4 Day 5 I session II session III session LeD RAM testing RAM testing RAM testing LeOS Morning Observation of: Observation of: RAM testing RAM testing RAM testing I trial: 1–2–3–4–5–6–7–8 VII trial: 7–8–1–2–3–4–5–6 II trial: 2–3–4–5–6–7–8–1 VIII trial: 8–1–2–3–4–5–6–7 III trial: 3–4–5–6–7–8–1–2 IX trial: 1–2–3–4–5–6–7–8 Afternoon Observation of: IV trial: 4–5–6–7–8–1–2–3 V trial: 5–6–7–8–1–2–3–4 VI trial: 6–7–8–1–2–3–4–5 LeOR Morning Observation of: Observation of: RAM testing RAM testing RAM testing I trial: 2–7–8–3–5–1–4–6 VII trial: 1–6–2–5–4–7–3–8 II trial: 8–3–2–5–7–4–1–6 VIII trial: 2–4–7–8–3–5–1–6 III trial: 3–5–8–7–1–4–6–2 IX trial: 5–8–3–4–6–2–7–1 Afternoon Observation of: IV trial: 6–8–3–4–7–2–5–1 V trial: 4–7–1–6–8–2–3–5 VI trial: 7–4–1–8–5–3–2–6 The strings of numbers indicate the sequence of visited arms performed by the experimenter in the both conditions of learning by observation. Note that the experimenter explored the Radial Arm Maze entering only the adjacent arms in Learning by Observation of a highly Structured explorative strategy (LeOS), while he did not follow an evident navigational strategy in Learning by Observation of Random explorative strategy (LeOR) (45,90, 135, 180, or 360) the child made in each trial Results divided by the number of angles made 9 100;—declara- tive mastery, calculated as the percentage of trials in which Motor accuracy the child stopped the search after collecting the eight rewards as if he/she knew the task was finished. All children similarly performed the agility course (total In examining maze drawings, we evaluated the type of time 36.6 ± 2.2 s; total score 5.5 ± 1.3). Namely, almost representation, an index rating the egocentricity/allocen- all children failed in shooting for goal, an ability acquired tricity of drawings using a 5-point Likert scale (from 1: relatively later, while all children successfully performed clear egocentricity, to 5: clear allocentricity). To objec- the slaloming. Table 2 shows the percentage of children tively assess this parameter in children’s drawings we who efficaciously performed each item of the motor asked a coder blind to RAM conditions and expert in accuracy task. A one-way ANCOVA failed to reveal any mental spatial representations and human navigation to statistical difference among the three experimental groups score each drawing according to its egocentricity/ in total time (F(2,31) = 0.28; p = 0.75; g = 0.02) and allocentricity. total score (F(2,31) = 0.18; p = 0.98; g = 0.001). Statistical analyses Drawing test of human figure The data were first tested for normality (Shapiro–Wilk’s Qualitative analysis of children’s drawings revealed that all test) and homoscedasticity (Levene’s test). All data were children drew details of human figure in accordance with presented as the mean ± SD and were analyzed by one- or their age. All children drew many body parts, inserting the two-way analyses of variance (ANCOVAs) with repeated hands and the feet on to the arms and legs. Not only they measures (session/angle) and with age and gender as drew the main body parts but they added more details covariates followed by post hoc multiple comparisons including hairs and clothing features. Typically, the when appropriate (Duncan’s test). youngest children of our study used single lines and the 123 Psychological Research (2018) 82:1212–1223 1217 Table 2 Percentage of children Somersaulting Balancing Jumping Slaloming Crawling Catching Shooting for goal of the three experimental conditions successfully LeD 92 83 92 100 92 83 17 performing each motor task of LeOS 91 91 82 100 100 73 18 the agility course LeOR 100 92 77 100 92 77 15 LeD Learning by Doing, LeOS Learning by Observation of a highly Structured explorative strategy, LeOR Learning by Observation of Random explorative strategy session (F(2,66) = 3.64; p = 0.03; g = 0.09) effects, while the interaction was not significant (F(4,66) = 0.7; p = 0.59; g = 0.04). Post hoc comparisons on group effect revealed that the children who had observed the actor solving the maze with a structured strategy (LeOS group) performed the task with a significantly lower number of entries in comparison to children who had never observed (LeD group) (p = 0.006). The children who had observed the actor solving the RAM with a random strategy (LeOR group) explored the maze making a number of entries similar to that of children belonging to LeD and LeOS groups (Fig. 4b; Table 3a). Errors A two-way ANCOVA (group 9 session) revealed sig- nificant group (F(2,31) = 5.29; p = 0.01; g = 0.25) and session (F(2,66) = 3.85; p = 0.02; g = 0.10) effects. Fig. 3 Drawings of human figure. Examples of drawings of children The interaction was not significant (F(4,66) = 0.68; randomly selected among groups (Machover’s test) p = 0.60; g = 0.04). Post hoc comparisons on group effect revealed that the children belonging to LeOS group oldest ones drew pairs of lines to represent arms and legs made a significantly lower number of errors in compar- (Fig. 3). Their correct acquisition and internalization of the ison to children who had not observed (LeD group) body image suggested a cognitive development appropriate (p = 0.004), while the children belonging to LeOR group for their age. made a similar number of errors to LeD and LeOS chil- dren (Fig. 4c; Table 3a). Radial maze Spatial span Total time A two-way ANCOVA (group 9 session) failed to reveal A two-way ANCOVA (group 9 session) revealed signifi- significant group (F(2,31) = 2.09; p = 0.14; g = 0.12) 2 2 cant group (F(2,31) = 4.59; p = 0.01; g = 0.23) and and session (F(2,66) = 2.45; p = 0.09; g = 0.07) effects, P P session (F(2,66) = 3.41; p = 0.04; g = 0.09) effects, but the interaction was significant (F(4,66) = 2.52; while the interaction was not significant (F(4,66) = 0.63; p = 0.04; g = 0.13). Post hoc comparisons on the inter- p = 0.64; g = 0.04). Post hoc comparisons on group action revealed that in the third session all children who effect revealed that the children who had observed the actor had observed the actor (LeOS and LeOR groups) had span (LeOS and LeOR groups) took less time than those values significantly higher than children who directly belonging to LeD group (at least p\ 0.04) (Fig. 4a; experienced the maze (LeD group) (at least p\ 0.04) Table 3a). (Fig. 4d; Table 3a). Entries Perseverations A two-way ANCOVA (group 9 session) revealed signifi- No child performed consecutive entries into the same arm cant group (F(2,31) = 4.55; p = 0.01; g = 0.22) and or into a fixed sequence of arms during RAM exploration. 123 1218 Psychological Research (2018) 82:1212–1223 Fig. 4 Performances in the Radial Arm Maze task. Data are Doing group, LeOS Learning by Observation of a highly Structured expressed as mean ± SD. The asterisks indicate the significance explorative strategy, LeOR, Learning by Observation of Random level of post hoc comparisons among groups (*p \ 0.05; explorative strategy group **p \ 0.01). In this and in the following figures: LeD Learning by Angle analysis Declarative mastery The angles performed in visiting RAM arms were closely A one-way ANCOVA was significant (F(2,31) = 5.75; linked to the navigational strategies put into action in p = 0.007; g = 0.41). Post hoc comparisons (LeD vs. exploring the maze. In the angle analysis, 360 angles are LeOS, p = 0.003; LeD vs. LeOR, p = 0.42; LeOS vs. missing because no child performed them. The experi- LeOR, p = 0.02) demonstrated that LeOS children mental procedure provided that the LeOS children obtained a significantly higher percentage of declarative observed the actor entering adjacent arms and making thus mastery in comparison to LeD and LeOR children (Fig. 6). only 45 angles, while LeOR children observed the actor performing 45 (14% of total angles), 90 (25%), 135 Drawing the maze (49%) and 180 (11%) angles (Fig. 5). A two-way ANCOVA (group 9 angle) failed to reveal a significant At the end of RAM testing, 12/12 LeD children, 11/11 group effect (F(2,31) = 0.44; p = 0.65; g = 0.03), while LeOS children, and 8/13 LeOR children made a drawing of angle effect (F(3,99) = 43.67; p\ 0.00001; g = 0.57) the setting where they had just played. The five uncoop- and interaction (F(6,99) = 4.45; p = 0.0005; g = 0.21) erative LeOR children who did not want to draw the maze were significant. Interestingly, post hoc comparisons on were not forced to do it. interaction demonstrated that LeOS children obtained a The type of representation of the experimental setting significantly higher percentage (74%) of 45 angles in was significantly different among groups (one-way comparison to others groups (LeD 46%; LeOR 55%; at ANCOVA (F(2,26) = 41.36; p \ 0.000001; g = 0.31; least p \ 0.01), and LeOR children obtained a significantly Post hoc comparisons: LeD vs. LeOS, p = 0.00006; LeD higher percentage (26%) of 135 angles in comparison to vs. LeOR, p = 0.0004; LeOS vs. LeOR, p = 0.0002). In others groups (LeD 7%; LeOS 6%; at least p\ 0.046) fact, the LeD children reached a mean score of (Fig. 5; Table 3b). 1.25 ± 0.45, indication that most of them drew the maze 123 Psychological Research (2018) 82:1212–1223 1219 Table 3 Post hoc comparisons Behavioral parameter Groups of the significant factors of ANCOVAs LeD vs. LeOS LeD vs. LeOR LeOS vs. LeOR p; Cohen’s d; r p; Cohen’s d; r p; Cohen’s d; r (a) Post hoc comparisons on the group effect of the two-way ANCOVAs Total time (mean of the 3 sessions) p = 0.0007 p = 0.04 p = 0.09 d = 0.99 d = 0.43 d =-0.68 r = 0.44 r = 0.21 r =-0.32 Entries (mean of the 3 sessions) p = 0.006 p = 0.21 p = 0.08 d = 0.95 d = 0.34 d =-0.70 r = 0.42 r = 0.17 r =-0.33 Errors (mean of the 3 sessions) p = 0.004 p = 0.12 p = 0.11 d = 0.99 d = 0.43 d =-0.65 r = 0.44 r = 0.21 r =-0.31 Spatial span (third session) p = 0.0004 p = 0.04 p = 0.12 d = 21.47 d = 20.72 d = 0.20 r = 20.59 r = 20.34 r = 0.10 Angle Groups LeD vs. LeOS LeD vs. LeOR LeOS vs. LeOR p; Cohen’s d; r p; Cohen’s d; r p; Cohen’s d; r (b) Post hoc comparisons on the interaction of the two-way ANCOVA 45 p = 0.0008 p = 0.29 p = 0.01 d = 21.09 d =-0.28 d = 0.74 r = 20.48 r =-0.14 r = 0.35 90 p = 0.13 p = 0.11 p = 0.87 d = 0.92 d = 1.2 d = 0.13 r = 0.42 r = 0.52 r = 0.07 135 p = 0.81 p = 0.04 p = 0.03 d = 0.45 d = 21.28 d = 21.41 r = 0.22 r = 20.54 r = 20.58 180 p = 0.19 p = 0.25 p = 0.82 d = 1.19 d = 0.93 d =-0.26 r = 0.51 r = 0.42 r =-0.13 Bold values are statistically significant with an overtly egocentric representation. Conversely, the other’s actions (Torriero et al., 2007; Menghini, Vicari, LeOS children reached a mean score of 4.64 ± 0.92. Mandolesi, & Petrosini, 2011; Foti et al., Interestingly, LeOR children reached a mean score of 2013, 2014, 2015). Although these abilities continue to 2.88 ± 1.25, an intermediate score indicating that the mature throughout life, they are already present in pre- observation of less structured navigational strategies did schoolers and young children (Mandolesi, Petrosini, not allow building an allocentric representation of the Menghini, Addona, & Vicari, 2009a; Rohbanfard & Pro- environment (Fig. 7). teau, 2011; Marshall & Meltzoff, 2014; Carr et al., 2015). Complex to-be-learned skills have generally an organi- zational structure that can be dissected into smaller units or Discussion types of behavior (i.e., extended or direct exploration) and the acquisition by observation of single exploratory Learning by observation requires attentive and mnesic strategies allows studying the learning power of specific functions, sequencing and planning abilities, anticipatory behavioral units (Graziano et al., 2002). Conversely, a expectation of effects, motor imagery, as well as engage- paradigm that involves actual experiential learning of ment in joint attention, and understanding and reproducing explorative strategies renders almost impossible the 123 1220 Psychological Research (2018) 82:1212–1223 Fig. 5 Observed angles vs. performed angles. Data are expressed as mean ± SD. The asterisks indicate the significance level of post hoc comparisons among groups (*p \ 0.05; ***p \ 0.001) In fact, in comparison to LeD children, LeOR children took less time to end the trial and obtained higher span values. Interestingly, the strategy the children observed influenced their exploration, as indicated by angle analysis. While the LeOS children observed the actor performing only 45 angles, the LeOR children observed the actor performing different angles (45,90, 135 and 180) but most fre- quently 135 angles. Remarkably, when actively exploring the RAM, LeOS children performed mainly 45 angles Fig. 6 Declarative mastery. Data are expressed as mean ± SD. The asterisks indicate the significance level of post hoc comparisons (74% of their total angles), and LeOR children mainly 135 among groups (**p \ 0.0005) angles (26%), evidencing thus that the observational training influenced the observers to apply the main strategy singling out of single behavioral units. Starting from these they had observed. It is worth noting that the tendency to perform 45 angles is the natural explorative pattern of premises, in the present study we singled out the obser- vational learning of different explorative strategies adopted healthy individuals in the RAM (Mandolesi et al., 2009a). In fact, the children of all experimental groups tended to in a constrained environment, such as the RAM, and we perform mainly 45 angles, although children of LeOS analyzed if the observation of the single navigational group performed the highest percentage of 45 angle. strategies could promote the acquisition of navigational On the basis of the present results it is possible to abilities and the building of the cognitive spatial map of the advance that behavioral units forming the strategy reper- explored environment in young children. toire employed in RAM exploration can be singularly The main result of the present research is that when the observed explorative procedure is structured, sequential acquired through observation. The children put into action the previously observed navigational strategy significantly and repetitive the action understanding and information storage processes are extremely effective. In fact, LeOS more frequently than the children who did not undergo any observational training (Fig. 5). In cognitive terms, this children made less entries and less errors, and reached learning could be described as a priming phenomenon, values of spatial span significantly higher than LeD chil- which increased the activation of stored internal represen- dren. However, also the observation of an unstructured and tations of a particular action. The primed records, now with random exploratory strategy facilitated RAM exploration. 123 Psychological Research (2018) 82:1212–1223 1221 Fig. 7 RAM representations. The drawings were made by the children of the three experimental groups at the end of the test increased salience, shaped the children’s successive and acquisition of navigational strategies (Leggio et al., exploratory behaviors. The observation of the actor’s 2000; Petrosini, 2007). In this regard, it was evidenced the behavior thus biased the observer’s pattern of behavior, activation of the cerebellar areas in many forms of the representing a real process of observational learning. ‘‘motor thought’’ whether or not it is accompanied by This interpretation is in agreement with the classic actual motor acts (Calvo-Merino, Grezes, Glaser, Pass- theoretical framework that posits that the observational ingham, & Haggard, 2006). learning requires that observers understand the other’s We wondered whether through observation of naviga- actions in terms of the same neural code they use to pro- tional strategies, the children really built a cognitive spatial duce the same motor behavior themselves (Decety & map or whether they learned to copy the observed trajec- Grezes, 1999) suggesting that the processes of learning by tories without developing any cognitive map. The explo- observation are very similar to the process of learning by rative behavior of the observer children was not a doing (Petrosini, 2007). stereotyped copy of the behaviors previously observed: The research on brain structures involved in observa- LeOS and LeOR children did not begin their exploration tional learning advances that the mirror neuron system that from the same arm explored as the first arm by the actor, is responsive to both observation and execution of action, they did not exhibit the same counter-clockwise or clock- may be differently integrated with other brain structures wise turning, and they did not exactly reproduce the depending on the kind of imitative task to be performed. sequence of entries. In short, they did not exhibit a mirror Namely, when observational learning is aimed at acquiring copy of the explorative behavior they had previously novel actions, activation of the mirror circuit may be observed. Their performances were coherent and elaborate integrated with the additional activation of the dorsolateral spatial procedures aimed at maze exploration. prefrontal cortex, an area correlated with the selection of Furthermore, it has to be taken into account that the motor acts, and with the activation of the premotor areas children were in a different spatial position during obser- relevant to motor preparation (Iacoboni, 2005). In the task vation and during testing, forcing them to allocentrically of the present research, the observational learning was encode the environmental coordinates. Notably, the aimed at developing efficient explorative strategies and observation of a specific navigational strategy helped the building cognitive spatial map. Probably, besides the pre- children to build the cognitive spatial map and conse- viously quoted cortical areas, the activation of the mirror quently to acquire/enrich the declarative knowledge of the system can be integrated with the activation of the cere- environment. Because of the not yet complete functional bellar areas known to be implicated in procedural learning maturation of the cerebral networks involved in spatial 123 1222 Psychological Research (2018) 82:1212–1223 information processing (Overman, Pate, Moore, & Peuster, References 1996; Lehnung et al., 1998; Leplow et al., 2003), the Bandura, A. (1977). Social learning theory. Englewood Cliffs, NJ: declarative spatial competence is not yet fully developed in Prentice Hall. children younger than about 7 years of age and reaches its Bird, G., & Heyes, C. (2005). Effector-dependent learning by complete development in late childhood and adolescence observation of a finger movement sequence. Journal of Exper- as the maturation of the fronto-parietal network occurs imental Psychology: Human Perception and Performance, 31, 262–275. (Klingberg, 2006). Since the children to the present Bjork, R. A. (2011). On the symbiosis of remembering, forgetting and research were aged 5.3 years on average, the processes learning. In A. S. Benjamin (Ed.), Successful remembering and underlying the acquisition of cognitive spatial map were successful forgetting: A Festschrift in honor of Robert A. Bjork. still immature. In fact, as it is typical of their age (Man- London: Psychology Press. Bjork, R. A., Dunlosky, J., & Kornell, N. (2013). Self-regulated dolesi et al., 2009b), the children of LeD and LeOR groups learning: Beliefs, techniques, and illusions. Annual Review of did not stop their search after collecting all rewards in a Psychology, 64, 417–444. higher percentage of trials in comparison to LeOS children Boncori, L. (2006). I test in Psicologia. Bologna: Il Mulino. the majority of whom (although had the same age) stopped Buchanan, J. J., & Dean, N. J. (2010). Specificity in practice benefits learning in novice models and variability in demonstration their exploration when they had collected the eight benefits observational practice. Psychological Research, 74, rewards. Interestingly, most drawings of the RAM of LeD 313–326. group were characterized by an egocentric vision of the Calvo-Merino, B., Gre `zes, J., Glaser, D. E., Passingham, R. E., & spatial context, in which the locations of the buckets were Haggard, P. (2006). Seeing or doing? Influence of visual and motor familiarity in action observation. Current Biology, 16, represented with respect to the particular perspective of the 1905–1910. child. LeOR children’s drawings represented the RAM Carr, K., Kendal, R. L., & Flynn, E. G. (2015). Imitate or innovate? with a midway vision between the egocentric and allo- Children’s innovation is influenced by the efficacy of observed centric extremes. Surprisingly, in most drawings of LeOS behaviour. Cognition, 142, 322–332. Caspers, S., Zilles, K., Laird, A. R., & Eickhoff, S. B. (2010). ALE children the RAM was depicted with a view from above, meta-analysis of action observation and imitation in the human with a clear allocentric perspective external to the child and brain. Neuroimage, 50, 1148–1167. independent of his or her position (Fig. 7). These findings Cox, M. (1992). Children’s drawings. London, England: Penguin indicate the capacity of the great majority of LeOS children Group. Decety, J., & Grezes, J. (1999). Neural mechanisms subserving the of transforming the egocentric information acquired during perception of human actions. Trends in Cognitive Science, 3, exploration in allocentric representation, and they support 172–178. the idea that to represent a new environment and build the Di Leo, J. (1970). Young children and their drawings. New York: cognitive spatial map a subject has to explore it appropri- Brunner/Mazel. Fagard, J., Rat-Fischer, L., Esseily, R., Somogyi, E., & O’Regan, J. K. ately (Mandolesi, Leggio, Spirito, & Petrosini, 2003). (2016). What does it take for an infant to learn how to use a tool by observation? Frontiers in Psychology, 1(7), 267. Acknowledgements We thank the children, parents, and Flynn, E., & Whiten, A. (2013). Dissecting children’s observational schoolteachers who made this study possible and we acknowledge the learning of complex actions through selective video displays. help by Adriano Accarino for testing some children. We are grateful Journal of Experimental Child Psychology, 116(2), 247–263. to Prof. Cecilia Guariglia and Prof. Liana Palermo for their competent Foti, F., Mazzone, L., Menghini, D., De Peppo, L., Federico, F., and kind help in analyzing and scoring the children’s drawings. This Postorino, V., et al. (2014). Learning by observation in children ´ ˆ work was supported by the Foundation Jerome Lejeune to L.M (no. with autism spectrum disorder. Psychological Medicine, 44, 886, 2011A; no. 1567, 2016B) and by PROGETTO CREME 2437–2447. ‘‘Campania Research in Experimental Medicine’’ POR Campania Foti, F., Menghini, D., Mandolesi, L., Federico, F., Vicari, S., & FSE 2007/2013 to P.B. Petrosini, L. (2013). Learning by observation: insights from Williams syndrome. PLoS One, 8(1), e53782. Compliance with ethical standards Foti, F., Menghini, D., Orlandi, E., Rufini, C., Crino ` , A., Spera, S., … Mandolesi, L. (2015). Learning by observation and learning by Conflict of interest The authors declare that they have no conflict of doing in Prader–Willi syndrome. Journal of Neurodevelopmental interest. Disorders, 7, 6. Frith, C. D., & Frith, U. (2012). Mechanisms of social cognition. Ethical standards All procedures performed in present study were in Annual Review of Psychology, 63, 287–313. accordance with the 1964 Helsinki declaration and its later amend- Gallese, V., Fadiga, L., Fogassi, L., & Rizzolatti, G. (1996). Action ments or comparable ethical standards. recognition in the premotor cortex. Brain, 119(2), 593–609. Graziano, A., Leggio, M. G., Mandolesi, L., Neri, P., Molinari, M., & Open Access This article is distributed under the terms of the Petrosini, L. (2002). Learning power of single behavioral units in Creative Commons Attribution 4.0 International License (http://crea acquisition of a complex spatial behavior: an observational tivecommons.org/licenses/by/4.0/), which permits unrestricted use, learning study in cerebellar-lesioned rats. Behavioral Neuro- distribution, and reproduction in any medium, provided you give science, 116, 116–125. appropriate credit to the original author(s) and the source, provide a Herold, K. H., & Akhtar, N. (2008). Imitative learning from a third- link to the Creative Commons license, and indicate if changes were party interaction: Relations with self-recognition and perspective made. 123 Psychological Research (2018) 82:1212–1223 1223 taking. Journal of Experimental Child Psychology, 101, psychology and cognitive neuroscience. Philosophical Transac- 114–123. tions of the Royal Society of London: Series B, Biological Iacoboni, M. (2005). Neural mechanisms of imitation. Current Sciences, 358, 491–500. Opinion Neurobiology, 15, 632–637. Meltzoff, A. N., Kuhl, P. K., Movellan, J., & Sejnowski, T. J. (2009). Keetch, K. M., Schmidt, R. A., Lee, T. D., & Young, D. E. (2005). Foundations for a new science of learning. Science, 325, Especial skills: their emergence with massive amounts of 284–288. practice. Journal of Experimental Psychology: Human Percep- Meltzoff, A. N., & Moore, M. K. (1977). Imitation of facial and tion and Performance, 31, 970–978. manual gestures by human neonates. Science, 198, 75–78. Klingberg, T. (2006). Development of a superior frontal-intraparietal Menghini, D., Vicari, S., Mandolesi, L., & Petrosini, L. (2011). Is network for visuo-spatial working memory. Neuropsychologia, learning by observation impaired in children with dyslexia? 44, 2171–2177. Neuropsychologia, 49, 1996–2003. Leggio, M. G., Molinari, M., Neri, P., Graziano, A., Mandolesi, L., & Molenberghs, P., Brander, C., Mattingley, J. B., & Cunnington, R. Petrosini, L. (2000). Representation of actions in rats: The role of (2010). The role of the superior temporal sulcus and the mirror cerebellum in learning spatial performances by observation. neuron system in imitation. Human Brain Mapping, 31(9), Proceedings of the National Academy of Science of United States 1316–1326. of America, 97, 2320–2325. Nadel, J. (2002). Imitation and imitation recognition: functional use in Lehnung, M., Leplow, B., Friege, L., Herzog, A., Ferstl, R., & preverbal infants and nonverbal children with autism. The Mehdorn, M. (1998). Development of spatial memory and Imitative Mind: Development, Evolution, and Brain Bases. In A. spatial orientation in preschoolers and primary school children. Meltzoff & W. Prinz (Eds.), The Imitative Mind (pp. 42–62). British Journal of Psychology, 89, 463–480. Cambridge: Cambridge University Press. Leplow, B., Lehnung, M., Pohl, J., Herzog, A., Ferstl, R., & Mehdorn, Nadel, J., & Butterworth, G. (1998). Imitation in infancy. Cambridge: M. (2003). Navigational place learning in children and young Cambridge University Press. adults as assessed with a standardized locomotor search task. Niederer, I., Kriemler, S., Gut, J., Hartmann, T., Schindler, C., Barral, British Journal of Psychology, 94, 299–317. J., & Puder, J. J. (2011). Relationship of aerobic fitness and Machover, K. (1949). Personality projection: In the drawing of a motor skills with memory and attention in preschoolers human figure. Springfield, IL: Charles C Thomas Publisher. (Ballabeina): A cross-sectional and longitudinal study. BioMed Mandolesi, L., Addona, F., Foti, F., Menghini, D., Petrosini, L., & Central Pediatrics, 11, 34. Vicari, S. (2009a). Spatial competences in Williams syndrome: a Overman, W. H., Pate, B. J., Moore, K., & Peuster, A. (1996). radial arm maze study. International Journal of Developmental Ontogeny of place learning in children as measured in the radial Neuroscience, 27, 205–213. arm maze, Morris search task, and open field task. Behavioral Mandolesi, L., Leggio, M. G., Spirito, F., & Petrosini, L. (2003). Neuroscience, 110, 1205–1228. Cerebellar contribution to spatial event processing: Do spatial Petrosini, L. (2007). ‘‘Do what I do’’ and ‘‘Do how I do’’: Different procedures contribute to formation of spatial declarative knowl- components of imitative learning are mediated by different edge? European Journal Neuroscience, 18, 2618–2626. neural structures. Neuroscientist, 13(4), 335–348. Mandolesi, L., Petrosini, L., Menghini, D., Addona, F., & Vicari, S. Rizzolatti, G., & Craighero, L. (2004). The mirror-neuron system. (2009b). Children’s radial arm maze performance as a function Annual Review of Neuroscience, 27, 169–192. of age and sex. International Journal of Developmental Neuro- Rizzolatti, G., Fogassi, L., & Gallese, V. (2001). Neurophysiological science, 27, 789–797. mechanisms underlying the understanding and imitation of Marshall, P. J., & Meltzoff, A. N. (2014). Neural mirroring action. Nature Reviews Neuroscience, 2(9), 661–670. mechanisms and imitation in human infants. Philosophical Rizzolatti, G., & Sinigaglia, C. (2010). The functional role of the transactions of the Royal Society of London Series B, Biological parieto-frontal mirror circuit: Interpretations and misinterpreta- sciences, 369, 2320–2325. tions. Nature Reviews Neuroscience, 11(4), 264–274. Matheson, H., Moore, C., & Akhtar, N. (2013). The development of Rohbanfard, H., & Proteau, L. (2011). Learning through observation: social learning in interactive and observational contexts. Journal A combination of expert and novice models favors learning. of Experimental Child Psychology, 114, 161–172. Experimental Brain Research, 215, 183–197. McGuigan, N. (2013). The influence of model status on the tendency Schmidt, R. A., & Bjork, R. A. (1992). New conceptualizations of of young children to over-imitate. Journal of Experimental Child practice: Common principles in three paradigms suggest new Psychology, 116, 962–969. concepts for training. Psychological Science, 3, 207–217. McGuigan, N., Gladstone, D., & Cook, L. (2012). Is the cultural Shimpi, P. M., Akhtar, N., & Moore, C. (2013). Toddlers’ imitative transmission of irrelevant tool actions in adult humans (Homo learning in interactive and observational contexts: The role of sapiens) best explained as the result of an evolved conformist age and familiarity of the model. Journal of Experimental Child bias? PLoS One, 7(12), e50863. Psychology, 116, 309–323. Meltzoff, A. N., & Andrew, N. (1995). Understanding the intentions Torriero, S., Oliveri, M., Koch, G., Caltagirone, C., & Petrosini, L. of others: Re-enactment of intended acts by 18-month-old (2007). The what and how of observational learning. Journal of children. Developmental Psychology, 31, 838–850. Cognitive Neuroscience, 19, 1656–1663. Meltzoff, A. N., & Decety, J. (2003). What imitation tells us about social cognition: A rapprochement between developmental http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Psychological Research Springer Journals

Are young children able to learn exploratory strategies by observation?

Download PDF
12 pages

Loading next page...
 
/lp/springer-journals/are-young-children-able-to-learn-exploratory-strategies-by-observation-GYx1A0wZMT

References (58)

Publisher
Springer Journals
Copyright
Copyright © 2017 by The Author(s)
Subject
Psychology; Psychology Research
ISSN
0340-0727
eISSN
1430-2772
DOI
10.1007/s00426-017-0896-0
pmid
28725993
Publisher site
See Article on Publisher Site

Abstract

Psychological Research (2018) 82:1212–1223 https://doi.org/10.1007/s00426-017-0896-0 O R I G IN AL ARTI CL E Are young children able to learn exploratory strategies by observation? 1,2 3 3,4 3 • • • • Francesca Foti Domenico Martone Stefania Orru` Simone Montuori 4 3,4 2,5 2,3,6 • • • Esther Imperlini Pasqualina Buono Laura Petrosini Laura Mandolesi Received: 21 December 2016 / Accepted: 14 July 2017 / Published online: 20 July 2017 The Author(s) 2017. This article is an open access publication Abstract New competencies may be learned through children who directly performed the RAM task without any active experience (experiential learning or learning by observation. The main result of the present research is that doing) or observation of others’ experiences (learning by the children who observed the highly structured and correct observation). Observing another person performing a exploratory strategy spent less time, made fewer errors, complex action facilitates the observer’s acquisition of the exhibited a longer spatial span, and thus they explored the same action. The present research is aimed at analyzing if maze more efficiently than the children who directly per- the observation of specific explorative strategies adopted in formed the RAM task without any observation. This finding a constrained environment, such as the Radial Arm Maze indicates that when the observed explorative procedure is (RAM), could help young children to explore the maze and structured, sequential and repetitive the action understand- to build a cognitive spatial map of the explored environ- ing and information storage processes are more effective. ment. To this aim young children were randomly assigned Importantly, the observation of specific spatial strategies to three groups: children who performed the RAM task helped the children to build the cognitive spatial map of the following the observation of an actor solving the same maze explored environment and consequently to acquire/enrich by putting into action a highly structured exploratory the declarative knowledge of the environment. strategy; children who performed the RAM task following the observation of the actor solving the same maze by putting into action a less structured exploratory strategy; Introduction New competencies may be learned through active experi- Francesca Foti and Domenico Martone contributed equally to this work. ence (experiential learning or learning by doing) or observation of others’ experiences (learning by observa- & Laura Mandolesi tion) (Bandura, 1977; Meltzoff, Kuhl, Movellan, & Sej- laura.mandolesi@uniparthenope.it nowski, 2009). Department of Medical and Surgical Sciences, University Learning by observation does not just involve copying ‘‘Magna Graecia’’, Catanzaro, Italy an action, but it requires that the observer transforms the IRCCS Fondazione Santa Lucia, Via del Fosso di Fiorano 65, observation into an action as similar as possible to the 00143 Rome, Italy model in terms of the goal to be reached and motor Department of Movement Sciences and Wellbeing, strategies to be applied (Meltzoff & Andrew, 1995; Melt- University ‘‘Parthenope’’, Naples, Italy zoff & Decety, 2003). Observing another person perform- ing a complex action represents a desirable condition of Fondazione IRCCS SDN, Naples, Italy learning that enables the learner to better understand the Department of Psychology, ‘‘Sapienza’’ University of Rome, skill prior to the performance and/or it helps the learner to Rome, Italy 6 more readily discriminate perceptually variables that are Department of Motor Science and Wellbeing, University important for the performance of that skill (Bird & Heyes, ‘‘Parthenope’’, Via Medina, 40, 80133 Naples, Italy 123 Psychological Research (2018) 82:1212–1223 1213 2005; Meltzoff et al., 2009). It is believed that observation solution. The findings of these researches suggest that of an action facilitates motor learning of that skill because young children are selective copiers who reproduce the it facilitates the acquisition of the main spatial and tem- irrelevant tool actions most frequently after having viewed poral features of the task, and thus removes the need to high-status models performing them. Thus, learning by create a cognitive representation of the action pattern observation represents a learning mechanism that can be through experiential learning (Keetch, Schmidt, Lee, & used in several fields (e.g., school and sport) as a ‘‘learning Young, 2005; Buchanan & Dean, 2010; Rohbanfard & technique’’. In addition, several studies have highlighted Proteau, 2011). However, it is worth of noting that condi- the importance of the observational learning in children tions of learning that accelerate the learning, by limiting with intellectual disabilities (Foti et al., 2013, 2014, 2015). the time-consuming process of learning by trial and error As for its neurobiological basis, the learning by obser- and reducing the practice needed to learn, often fail to vation is thought to utilize brain regions responsive to both support long-term retention and transfer (Schmidt & Bjork, observation and execution of action, as the mirror neuron 1992; Bjork, 2011; Bjork, Dunlosky, & Kornell, 2013). system (Gallese, Fadiga, Fogassi, & Rizzolatti, 1996; Acquiring skills by observation is a fundamental cog- Rizzolatti, Fogassi, & Gallese, 2001; Rizzolatti & nitive ability already existing from the birth (Meltzoff & Craighero, 2004). The mirror neuron system includes pre- Moore, 1977; Nadel & Butterworth, 1998; Meltzoff et al., motor cortex, inferior frontal gyrus, and inferior parietal 2009; Nadel, 2002). Already at 18-months-old children lobule, areas which receive their main visual input from the may learn a novel motor pattern by observation (Herold & superior temporal sulcus (Molenberghs, Brander, Mattin- Akhtar, 2008; Matheson, Moore, & Akhtar, 2013) and if gley, & Cunnington, 2010; Caspers, Zilles, Laird, & the adults explicitly show their intention prior to demon- Eickhoff, 2010). Insofar as it generates a simulation circuit stration, even 16-months-old infants learn by observation that allows the association between one’s own actions with (Fagard, Rat-Fischer, Esseily, Somogyi, & O’Regan, others’ actions, the mirror neuron system is retained to be 2016). Three-year-old children are able to learn how to involved in action understanding, imagination, and imita- extract a reward from a box following a video-demon- tion (Rizzolatti & Sinigaglia, 2010), and thus even in the stration of the correct procedure (Flynn & Whiten, 2013). observational learning. Besides imitative abilities learning by observation Most developmental studies focused on how and what requires cognitive competencies, as attentive and mnesic the observer child has to observe to promote learning functions, sequencing abilities, planning, response inhibi- (Rohbanfard & Proteau, 2011; Marshall & Meltzoff, 2014; tion, cognitive flexibility, good knowledge and anticipatory Carr et al., 2015). However, to our knowledge there are no expectation of effects related to actions, goal-directed developmental studies that investigated whether the actions, and motor imagery allowing recombination of learning by observation of exploratory strategies promotes novel actions with novel effects (Foti et al., the acquisition of navigational abilities. For this reason, we 2013, 2014, 2015; Torriero, Oliveri, Koch, Caltagirone, & wondered if observing an adult actor who adopts specific Petrosini, 2007). Furthermore, to learn by observation it is navigational strategies to explore a radial arm maze (RAM) necessary to observe and attend to the actor, engage in joint can help young children improve their exploration of the attention, understand and reproduce other’s actions. Thus, same maze and build the cognitive spatial map of the learning by observation also represents a powerful social explored environment. Another aim of the present research learning mechanism (Frith & Frith, 2012). For example, is to determine whether observation of a structured model children can learn how to behave in social contexts by or of a less structured model of explorative strategies observing how adults interact with each other (Shimpi, would have resulted in different reproduction of the Akhtar, & Moore, 2013). Recently, it was shown that if the explorative patterns. On one hand, it has been proposed model has a high social status, such as a teacher, the that the observation permits the observer to develop a sort children tend to learn even irrelevant information by of ‘‘perceptual blueprint’’ of the task to be learned (Ban- observation (McGuigan, Gladstone, & Cook, 2012; dura, 1977). This may work in favor of the utilization of a McGuigan, 2013) or attempts without outcome (Carr, model performing a very efficient and successful explo- Kendal, & Flynn, 2015). The typical scenario in these rative strategy. On the other hand, it might be fruitful also studies is that before being allowed to attempt the task to observe a model performing a less structured and with a themselves, the observers watch an adult model perform a superior mnesic load, but still successful explorative sequence of tool actions varying according to their causal strategy. To these aims, young children (mean age: 5 years necessity, with some of the actions being necessary for and 3 months) were randomly assigned to three groups: reward retrieval, others being causally irrelevant (as per- children who performed the RAM task following the forming unnecessary taps before retrieving a reward from a observation of an actor solving the same maze by putting box) and others without the efficacy of an observed into action a highly structured and efficient exploratory 123 1214 Psychological Research (2018) 82:1212–1223 strategy (such as sequentially entering adjacent arms); children who performed the RAM task following the observation of the actor solving the same maze by putting into action a less structured and still successful exploratory strategy (such as randomly entering arms); children who directly performed the RAM task without any observa- tional training. Methods Participants Thirty-six healthy Italian children (17 M and 19 F) aged from 4 years and 6 months (4.6) to 5 years and 9 months (5.9) (mean age ± SD 5.3 ± 0.2) participated in the pre- sent study. Children were subdivided into three groups according to the following experimental conditions: Learning by Observation of a highly Structured explorative Fig. 1 Schematic representation of the agility course strategy (LeOS) (N = 11; 5 M and 6 F; mean age 5.2 ± 0.3); Learning by Observation of Random explo- rative strategy (LeOR) (N = 13; 6 M and 7 F; mean age Slaloming: the child runs in a zig zag pattern among six cones; Crawling: the child grovels under a rod held by two 5.3 ± 0.2); Learning by Doing (LeD) (N = 12; 6 M and 6 F; mean age 5.3 ± 0.1). No children have had previous cones set at 50 cm (19.68 in) from the ground; Catching: the child enters a circle placed on the ground in which he/ experience with the RAM task. All children had normal or corrected-to-normal vision she grasps a ball thrown by the teacher positioned in front of him/her; Shooting for goal: the child throws the ball into and standard anthropometric measurements and presented the basket located in front of him/her. no neurological or neuropsychological problems. Body Total score (the sum of scores ranging from 0 to 7) and mass index values (M = 17.06 ± 2.8; F = 16.83 ± 1.62) total time (time to perform the entire course) were were between the 50th–75th percentile. To exclude the recorded. presence of sensory-motor deficits, the psychomotor development of all children was evaluated through a bat- tery of exercises of motor accuracy (Niederer et al., 2011). Drawing test of the human figure To verify graphic abilities and cognitive development, all children were assessed in the drawing test of the human According to Machover’s instructions (Machover, 1949), each child was asked to ‘‘draw someone’’. For child’s figure (Machover, 1949). All children attended a kinder- question on what it was possible to draw, the experimenter garten school in South Italy where a 1 h/day of physical replied ‘‘whatever you want’’. If the child drew only the activity was planned for 5 days/week. The parents of head, the investigator encouraged him/her to draw the children gave informed written consent. The study was whole figure. Since the children were less than 6 years of conducted according to the 1964 Declaration of Helsinki. age, the qualitative assessment of drawing human fig- Motor accuracy assessment ure was focused to highlight whether the child did not draw significant details, such as hands, hair, eyes, mouth (Di In the school gym an ‘‘agility course’’ was built (Fig. 1) Leo, 1970; Cox, 1992; Boncori, 2006). where the following seven motor abilities were evaluated assigning ‘‘1’’ or ‘‘0’’ scores according to correctness or Apparatus incorrectness, respectively. The RAM adapted for children consisted of a round central Somersaulting: the child rolls forward in a complete platform [1 m (39.37 in) in diameter] with eight arms revolution around the horizontal axis on a carpet at the start [50 cm (19.68 in) wide 9 3.5 m (137.79 in) long] radiating of the course; Balancing: the child walks heel to toe on a like the spokes of a wheel (Fig. 2). To force the child to white 2 m (78.74 in) tape (10 cm (3.93 in) large) fixed on exit from an arm and return to the center of the starting to the ground; Jumping: the child hops three 25 cm (9.84 in) high obstacles, built with two cones joined by a rod; platform before entering another arm, the sides of each arm 123 Psychological Research (2018) 82:1212–1223 1215 bucket only once. An error was made when the child re- entered an arm already visited during the same trial. Each child performed three trials a day for three consecutive days. Since the three daily trials constituted a session and each child made three sessions, each child performed nine trials. At the end of each trial, the child waited 1 h (inter- trial interval), before being re-tested in the RAM. At the beginning of RAM testing, the experimenter used the same simple verbal instructions to explain the task to each child (‘‘The game is to find the little colored balls. Do you see the colored buckets at the end of each alley? You have to reach a bucket, take the little ball inside, and then go back to the center, where the platform is, until you have col- Fig. 2 View of the eight-arm radial maze lected all the balls. Be careful to reach the buckets always were marked off by white and red ribbons hung across the staying inside the maze. Go and have fun!’’). No other opening and the end of the arm, forming a sort of con- instructions or verbal encouragement were provided during straining barrier. This procedure prevented the children testing. In the two observation conditions (LeOS and from ‘‘cutting corners’’ as they exited from an arm and LeOR), before starting RAM exploration the experimenter forced them to exit and return to the center of the starting told the children: ‘‘The game is to find the little colored platform before entering another arm. At the end of each balls inside the buckets. Look at me carefully’’. In the arm, there was a red plastic bucket (18 cm (7.08 in) LeOS condition, each child observed three sessions of three wide 9 28 cm (11.02 in) high) containing the reward (a trials each in which the actor explored the RAM entering little colored ball). The RAM, located outdoors in a foot- always the adjacent arms and stopped after the eight ball field, was surrounded by extra-maze cues (trees, rewards were collected. In the LeOR condition, each child swings, benches, etc.) held in constant spatial relations observed three sessions of three trials each in which the throughout the experiment. The arms were virtually num- actor explored the RAM using a pseudorandom explorative bered in a clockwise direction, considering arm 1 as the strategy and stopping after eight rewards collected. In farthest from the experimenter’s location. Only during the LeOS and LeOR conditions, children observed the actor at experiment could the children see the maze or have phys- distance of about 1.5 m (59.05 in) from the RAM, chang- ical access to it. To increase the motivation of picking up ing their point of observation at every session. Then, each the rewards, at the end of each trial the child received a child actively experienced the three RAM sessions (RAM reward (a little toy) in exchange for all the colored balls testing; Table 1). The trials were annulled if the child left found in the buckets. the maze. However, very few children of LeD condition engaged in this behavior and, in any case, only in the very Experimental procedure first trials of the task. In LeOS and LeOR conditions, no child left the maze. The RAM testing lasted 3 consecutive The three experimental conditions were:—Learning by days and in this execution phase all children were video- Observation of a highly Structured explorative strategy taped and recorded manually. At the end of RAM testing (LeOS), in which the children performed the RAM task phase, all children were asked to make a drawing of the following the observation of an actor solving the maze setting where they had just ‘‘played’’ to evaluate their through a highly structured exploratory strategy;—Learn- mental representative mapping abilities. ing by Observation of Random explorative strategy (LeOR), in which the children performed the RAM task Behavioral parameters following the observation of the actor solving the maze through a less structured exploratory strategy;—Learning We evaluated:—total time (in seconds) spent to complete by Doing (LeD) in which the children directly performed the task;—entries, calculated as the number of visited the RAM task without observation (Table 1). arms;—errors, calculated as the number of re-entries into In each trial of the RAM testing, each child was allowed already visited arms;—spatial span, calculated as the to explore freely the eight arms to retrieve the reward. A longest sequence of correctly visited arms;—persevera- trial ended when all eight rewards had been collected, 20 tions, calculated as the percentage of consecutive entries choices had been made, or 10 min had elapsed from the into the same arm or the re-entries into a fixed sequence of start of the task. Since the buckets were never rewarded arms, divided by the number of arms visited;—percentage twice, the optimal performance consisted of visiting each of angled turns, calculated as the number of a given angle 123 1216 Psychological Research (2018) 82:1212–1223 Table 1 Experimental procedures of the three experimental conditions Day 1 Day 2 Day 3 Day 4 Day 5 I session II session III session LeD RAM testing RAM testing RAM testing LeOS Morning Observation of: Observation of: RAM testing RAM testing RAM testing I trial: 1–2–3–4–5–6–7–8 VII trial: 7–8–1–2–3–4–5–6 II trial: 2–3–4–5–6–7–8–1 VIII trial: 8–1–2–3–4–5–6–7 III trial: 3–4–5–6–7–8–1–2 IX trial: 1–2–3–4–5–6–7–8 Afternoon Observation of: IV trial: 4–5–6–7–8–1–2–3 V trial: 5–6–7–8–1–2–3–4 VI trial: 6–7–8–1–2–3–4–5 LeOR Morning Observation of: Observation of: RAM testing RAM testing RAM testing I trial: 2–7–8–3–5–1–4–6 VII trial: 1–6–2–5–4–7–3–8 II trial: 8–3–2–5–7–4–1–6 VIII trial: 2–4–7–8–3–5–1–6 III trial: 3–5–8–7–1–4–6–2 IX trial: 5–8–3–4–6–2–7–1 Afternoon Observation of: IV trial: 6–8–3–4–7–2–5–1 V trial: 4–7–1–6–8–2–3–5 VI trial: 7–4–1–8–5–3–2–6 The strings of numbers indicate the sequence of visited arms performed by the experimenter in the both conditions of learning by observation. Note that the experimenter explored the Radial Arm Maze entering only the adjacent arms in Learning by Observation of a highly Structured explorative strategy (LeOS), while he did not follow an evident navigational strategy in Learning by Observation of Random explorative strategy (LeOR) (45,90, 135, 180, or 360) the child made in each trial Results divided by the number of angles made 9 100;—declara- tive mastery, calculated as the percentage of trials in which Motor accuracy the child stopped the search after collecting the eight rewards as if he/she knew the task was finished. All children similarly performed the agility course (total In examining maze drawings, we evaluated the type of time 36.6 ± 2.2 s; total score 5.5 ± 1.3). Namely, almost representation, an index rating the egocentricity/allocen- all children failed in shooting for goal, an ability acquired tricity of drawings using a 5-point Likert scale (from 1: relatively later, while all children successfully performed clear egocentricity, to 5: clear allocentricity). To objec- the slaloming. Table 2 shows the percentage of children tively assess this parameter in children’s drawings we who efficaciously performed each item of the motor asked a coder blind to RAM conditions and expert in accuracy task. A one-way ANCOVA failed to reveal any mental spatial representations and human navigation to statistical difference among the three experimental groups score each drawing according to its egocentricity/ in total time (F(2,31) = 0.28; p = 0.75; g = 0.02) and allocentricity. total score (F(2,31) = 0.18; p = 0.98; g = 0.001). Statistical analyses Drawing test of human figure The data were first tested for normality (Shapiro–Wilk’s Qualitative analysis of children’s drawings revealed that all test) and homoscedasticity (Levene’s test). All data were children drew details of human figure in accordance with presented as the mean ± SD and were analyzed by one- or their age. All children drew many body parts, inserting the two-way analyses of variance (ANCOVAs) with repeated hands and the feet on to the arms and legs. Not only they measures (session/angle) and with age and gender as drew the main body parts but they added more details covariates followed by post hoc multiple comparisons including hairs and clothing features. Typically, the when appropriate (Duncan’s test). youngest children of our study used single lines and the 123 Psychological Research (2018) 82:1212–1223 1217 Table 2 Percentage of children Somersaulting Balancing Jumping Slaloming Crawling Catching Shooting for goal of the three experimental conditions successfully LeD 92 83 92 100 92 83 17 performing each motor task of LeOS 91 91 82 100 100 73 18 the agility course LeOR 100 92 77 100 92 77 15 LeD Learning by Doing, LeOS Learning by Observation of a highly Structured explorative strategy, LeOR Learning by Observation of Random explorative strategy session (F(2,66) = 3.64; p = 0.03; g = 0.09) effects, while the interaction was not significant (F(4,66) = 0.7; p = 0.59; g = 0.04). Post hoc comparisons on group effect revealed that the children who had observed the actor solving the maze with a structured strategy (LeOS group) performed the task with a significantly lower number of entries in comparison to children who had never observed (LeD group) (p = 0.006). The children who had observed the actor solving the RAM with a random strategy (LeOR group) explored the maze making a number of entries similar to that of children belonging to LeD and LeOS groups (Fig. 4b; Table 3a). Errors A two-way ANCOVA (group 9 session) revealed sig- nificant group (F(2,31) = 5.29; p = 0.01; g = 0.25) and session (F(2,66) = 3.85; p = 0.02; g = 0.10) effects. Fig. 3 Drawings of human figure. Examples of drawings of children The interaction was not significant (F(4,66) = 0.68; randomly selected among groups (Machover’s test) p = 0.60; g = 0.04). Post hoc comparisons on group effect revealed that the children belonging to LeOS group oldest ones drew pairs of lines to represent arms and legs made a significantly lower number of errors in compar- (Fig. 3). Their correct acquisition and internalization of the ison to children who had not observed (LeD group) body image suggested a cognitive development appropriate (p = 0.004), while the children belonging to LeOR group for their age. made a similar number of errors to LeD and LeOS chil- dren (Fig. 4c; Table 3a). Radial maze Spatial span Total time A two-way ANCOVA (group 9 session) failed to reveal A two-way ANCOVA (group 9 session) revealed signifi- significant group (F(2,31) = 2.09; p = 0.14; g = 0.12) 2 2 cant group (F(2,31) = 4.59; p = 0.01; g = 0.23) and and session (F(2,66) = 2.45; p = 0.09; g = 0.07) effects, P P session (F(2,66) = 3.41; p = 0.04; g = 0.09) effects, but the interaction was significant (F(4,66) = 2.52; while the interaction was not significant (F(4,66) = 0.63; p = 0.04; g = 0.13). Post hoc comparisons on the inter- p = 0.64; g = 0.04). Post hoc comparisons on group action revealed that in the third session all children who effect revealed that the children who had observed the actor had observed the actor (LeOS and LeOR groups) had span (LeOS and LeOR groups) took less time than those values significantly higher than children who directly belonging to LeD group (at least p\ 0.04) (Fig. 4a; experienced the maze (LeD group) (at least p\ 0.04) Table 3a). (Fig. 4d; Table 3a). Entries Perseverations A two-way ANCOVA (group 9 session) revealed signifi- No child performed consecutive entries into the same arm cant group (F(2,31) = 4.55; p = 0.01; g = 0.22) and or into a fixed sequence of arms during RAM exploration. 123 1218 Psychological Research (2018) 82:1212–1223 Fig. 4 Performances in the Radial Arm Maze task. Data are Doing group, LeOS Learning by Observation of a highly Structured expressed as mean ± SD. The asterisks indicate the significance explorative strategy, LeOR, Learning by Observation of Random level of post hoc comparisons among groups (*p \ 0.05; explorative strategy group **p \ 0.01). In this and in the following figures: LeD Learning by Angle analysis Declarative mastery The angles performed in visiting RAM arms were closely A one-way ANCOVA was significant (F(2,31) = 5.75; linked to the navigational strategies put into action in p = 0.007; g = 0.41). Post hoc comparisons (LeD vs. exploring the maze. In the angle analysis, 360 angles are LeOS, p = 0.003; LeD vs. LeOR, p = 0.42; LeOS vs. missing because no child performed them. The experi- LeOR, p = 0.02) demonstrated that LeOS children mental procedure provided that the LeOS children obtained a significantly higher percentage of declarative observed the actor entering adjacent arms and making thus mastery in comparison to LeD and LeOR children (Fig. 6). only 45 angles, while LeOR children observed the actor performing 45 (14% of total angles), 90 (25%), 135 Drawing the maze (49%) and 180 (11%) angles (Fig. 5). A two-way ANCOVA (group 9 angle) failed to reveal a significant At the end of RAM testing, 12/12 LeD children, 11/11 group effect (F(2,31) = 0.44; p = 0.65; g = 0.03), while LeOS children, and 8/13 LeOR children made a drawing of angle effect (F(3,99) = 43.67; p\ 0.00001; g = 0.57) the setting where they had just played. The five uncoop- and interaction (F(6,99) = 4.45; p = 0.0005; g = 0.21) erative LeOR children who did not want to draw the maze were significant. Interestingly, post hoc comparisons on were not forced to do it. interaction demonstrated that LeOS children obtained a The type of representation of the experimental setting significantly higher percentage (74%) of 45 angles in was significantly different among groups (one-way comparison to others groups (LeD 46%; LeOR 55%; at ANCOVA (F(2,26) = 41.36; p \ 0.000001; g = 0.31; least p \ 0.01), and LeOR children obtained a significantly Post hoc comparisons: LeD vs. LeOS, p = 0.00006; LeD higher percentage (26%) of 135 angles in comparison to vs. LeOR, p = 0.0004; LeOS vs. LeOR, p = 0.0002). In others groups (LeD 7%; LeOS 6%; at least p\ 0.046) fact, the LeD children reached a mean score of (Fig. 5; Table 3b). 1.25 ± 0.45, indication that most of them drew the maze 123 Psychological Research (2018) 82:1212–1223 1219 Table 3 Post hoc comparisons Behavioral parameter Groups of the significant factors of ANCOVAs LeD vs. LeOS LeD vs. LeOR LeOS vs. LeOR p; Cohen’s d; r p; Cohen’s d; r p; Cohen’s d; r (a) Post hoc comparisons on the group effect of the two-way ANCOVAs Total time (mean of the 3 sessions) p = 0.0007 p = 0.04 p = 0.09 d = 0.99 d = 0.43 d =-0.68 r = 0.44 r = 0.21 r =-0.32 Entries (mean of the 3 sessions) p = 0.006 p = 0.21 p = 0.08 d = 0.95 d = 0.34 d =-0.70 r = 0.42 r = 0.17 r =-0.33 Errors (mean of the 3 sessions) p = 0.004 p = 0.12 p = 0.11 d = 0.99 d = 0.43 d =-0.65 r = 0.44 r = 0.21 r =-0.31 Spatial span (third session) p = 0.0004 p = 0.04 p = 0.12 d = 21.47 d = 20.72 d = 0.20 r = 20.59 r = 20.34 r = 0.10 Angle Groups LeD vs. LeOS LeD vs. LeOR LeOS vs. LeOR p; Cohen’s d; r p; Cohen’s d; r p; Cohen’s d; r (b) Post hoc comparisons on the interaction of the two-way ANCOVA 45 p = 0.0008 p = 0.29 p = 0.01 d = 21.09 d =-0.28 d = 0.74 r = 20.48 r =-0.14 r = 0.35 90 p = 0.13 p = 0.11 p = 0.87 d = 0.92 d = 1.2 d = 0.13 r = 0.42 r = 0.52 r = 0.07 135 p = 0.81 p = 0.04 p = 0.03 d = 0.45 d = 21.28 d = 21.41 r = 0.22 r = 20.54 r = 20.58 180 p = 0.19 p = 0.25 p = 0.82 d = 1.19 d = 0.93 d =-0.26 r = 0.51 r = 0.42 r =-0.13 Bold values are statistically significant with an overtly egocentric representation. Conversely, the other’s actions (Torriero et al., 2007; Menghini, Vicari, LeOS children reached a mean score of 4.64 ± 0.92. Mandolesi, & Petrosini, 2011; Foti et al., Interestingly, LeOR children reached a mean score of 2013, 2014, 2015). Although these abilities continue to 2.88 ± 1.25, an intermediate score indicating that the mature throughout life, they are already present in pre- observation of less structured navigational strategies did schoolers and young children (Mandolesi, Petrosini, not allow building an allocentric representation of the Menghini, Addona, & Vicari, 2009a; Rohbanfard & Pro- environment (Fig. 7). teau, 2011; Marshall & Meltzoff, 2014; Carr et al., 2015). Complex to-be-learned skills have generally an organi- zational structure that can be dissected into smaller units or Discussion types of behavior (i.e., extended or direct exploration) and the acquisition by observation of single exploratory Learning by observation requires attentive and mnesic strategies allows studying the learning power of specific functions, sequencing and planning abilities, anticipatory behavioral units (Graziano et al., 2002). Conversely, a expectation of effects, motor imagery, as well as engage- paradigm that involves actual experiential learning of ment in joint attention, and understanding and reproducing explorative strategies renders almost impossible the 123 1220 Psychological Research (2018) 82:1212–1223 Fig. 5 Observed angles vs. performed angles. Data are expressed as mean ± SD. The asterisks indicate the significance level of post hoc comparisons among groups (*p \ 0.05; ***p \ 0.001) In fact, in comparison to LeD children, LeOR children took less time to end the trial and obtained higher span values. Interestingly, the strategy the children observed influenced their exploration, as indicated by angle analysis. While the LeOS children observed the actor performing only 45 angles, the LeOR children observed the actor performing different angles (45,90, 135 and 180) but most fre- quently 135 angles. Remarkably, when actively exploring the RAM, LeOS children performed mainly 45 angles Fig. 6 Declarative mastery. Data are expressed as mean ± SD. The asterisks indicate the significance level of post hoc comparisons (74% of their total angles), and LeOR children mainly 135 among groups (**p \ 0.0005) angles (26%), evidencing thus that the observational training influenced the observers to apply the main strategy singling out of single behavioral units. Starting from these they had observed. It is worth noting that the tendency to perform 45 angles is the natural explorative pattern of premises, in the present study we singled out the obser- vational learning of different explorative strategies adopted healthy individuals in the RAM (Mandolesi et al., 2009a). In fact, the children of all experimental groups tended to in a constrained environment, such as the RAM, and we perform mainly 45 angles, although children of LeOS analyzed if the observation of the single navigational group performed the highest percentage of 45 angle. strategies could promote the acquisition of navigational On the basis of the present results it is possible to abilities and the building of the cognitive spatial map of the advance that behavioral units forming the strategy reper- explored environment in young children. toire employed in RAM exploration can be singularly The main result of the present research is that when the observed explorative procedure is structured, sequential acquired through observation. The children put into action the previously observed navigational strategy significantly and repetitive the action understanding and information storage processes are extremely effective. In fact, LeOS more frequently than the children who did not undergo any observational training (Fig. 5). In cognitive terms, this children made less entries and less errors, and reached learning could be described as a priming phenomenon, values of spatial span significantly higher than LeD chil- which increased the activation of stored internal represen- dren. However, also the observation of an unstructured and tations of a particular action. The primed records, now with random exploratory strategy facilitated RAM exploration. 123 Psychological Research (2018) 82:1212–1223 1221 Fig. 7 RAM representations. The drawings were made by the children of the three experimental groups at the end of the test increased salience, shaped the children’s successive and acquisition of navigational strategies (Leggio et al., exploratory behaviors. The observation of the actor’s 2000; Petrosini, 2007). In this regard, it was evidenced the behavior thus biased the observer’s pattern of behavior, activation of the cerebellar areas in many forms of the representing a real process of observational learning. ‘‘motor thought’’ whether or not it is accompanied by This interpretation is in agreement with the classic actual motor acts (Calvo-Merino, Grezes, Glaser, Pass- theoretical framework that posits that the observational ingham, & Haggard, 2006). learning requires that observers understand the other’s We wondered whether through observation of naviga- actions in terms of the same neural code they use to pro- tional strategies, the children really built a cognitive spatial duce the same motor behavior themselves (Decety & map or whether they learned to copy the observed trajec- Grezes, 1999) suggesting that the processes of learning by tories without developing any cognitive map. The explo- observation are very similar to the process of learning by rative behavior of the observer children was not a doing (Petrosini, 2007). stereotyped copy of the behaviors previously observed: The research on brain structures involved in observa- LeOS and LeOR children did not begin their exploration tional learning advances that the mirror neuron system that from the same arm explored as the first arm by the actor, is responsive to both observation and execution of action, they did not exhibit the same counter-clockwise or clock- may be differently integrated with other brain structures wise turning, and they did not exactly reproduce the depending on the kind of imitative task to be performed. sequence of entries. In short, they did not exhibit a mirror Namely, when observational learning is aimed at acquiring copy of the explorative behavior they had previously novel actions, activation of the mirror circuit may be observed. Their performances were coherent and elaborate integrated with the additional activation of the dorsolateral spatial procedures aimed at maze exploration. prefrontal cortex, an area correlated with the selection of Furthermore, it has to be taken into account that the motor acts, and with the activation of the premotor areas children were in a different spatial position during obser- relevant to motor preparation (Iacoboni, 2005). In the task vation and during testing, forcing them to allocentrically of the present research, the observational learning was encode the environmental coordinates. Notably, the aimed at developing efficient explorative strategies and observation of a specific navigational strategy helped the building cognitive spatial map. Probably, besides the pre- children to build the cognitive spatial map and conse- viously quoted cortical areas, the activation of the mirror quently to acquire/enrich the declarative knowledge of the system can be integrated with the activation of the cere- environment. Because of the not yet complete functional bellar areas known to be implicated in procedural learning maturation of the cerebral networks involved in spatial 123 1222 Psychological Research (2018) 82:1212–1223 information processing (Overman, Pate, Moore, & Peuster, References 1996; Lehnung et al., 1998; Leplow et al., 2003), the Bandura, A. (1977). Social learning theory. Englewood Cliffs, NJ: declarative spatial competence is not yet fully developed in Prentice Hall. children younger than about 7 years of age and reaches its Bird, G., & Heyes, C. (2005). Effector-dependent learning by complete development in late childhood and adolescence observation of a finger movement sequence. Journal of Exper- as the maturation of the fronto-parietal network occurs imental Psychology: Human Perception and Performance, 31, 262–275. (Klingberg, 2006). Since the children to the present Bjork, R. A. (2011). On the symbiosis of remembering, forgetting and research were aged 5.3 years on average, the processes learning. In A. S. Benjamin (Ed.), Successful remembering and underlying the acquisition of cognitive spatial map were successful forgetting: A Festschrift in honor of Robert A. Bjork. still immature. In fact, as it is typical of their age (Man- London: Psychology Press. Bjork, R. A., Dunlosky, J., & Kornell, N. (2013). Self-regulated dolesi et al., 2009b), the children of LeD and LeOR groups learning: Beliefs, techniques, and illusions. Annual Review of did not stop their search after collecting all rewards in a Psychology, 64, 417–444. higher percentage of trials in comparison to LeOS children Boncori, L. (2006). I test in Psicologia. Bologna: Il Mulino. the majority of whom (although had the same age) stopped Buchanan, J. J., & Dean, N. J. (2010). Specificity in practice benefits learning in novice models and variability in demonstration their exploration when they had collected the eight benefits observational practice. Psychological Research, 74, rewards. Interestingly, most drawings of the RAM of LeD 313–326. group were characterized by an egocentric vision of the Calvo-Merino, B., Gre `zes, J., Glaser, D. E., Passingham, R. E., & spatial context, in which the locations of the buckets were Haggard, P. (2006). Seeing or doing? Influence of visual and motor familiarity in action observation. Current Biology, 16, represented with respect to the particular perspective of the 1905–1910. child. LeOR children’s drawings represented the RAM Carr, K., Kendal, R. L., & Flynn, E. G. (2015). Imitate or innovate? with a midway vision between the egocentric and allo- Children’s innovation is influenced by the efficacy of observed centric extremes. Surprisingly, in most drawings of LeOS behaviour. Cognition, 142, 322–332. Caspers, S., Zilles, K., Laird, A. R., & Eickhoff, S. B. (2010). ALE children the RAM was depicted with a view from above, meta-analysis of action observation and imitation in the human with a clear allocentric perspective external to the child and brain. Neuroimage, 50, 1148–1167. independent of his or her position (Fig. 7). These findings Cox, M. (1992). Children’s drawings. London, England: Penguin indicate the capacity of the great majority of LeOS children Group. Decety, J., & Grezes, J. (1999). Neural mechanisms subserving the of transforming the egocentric information acquired during perception of human actions. Trends in Cognitive Science, 3, exploration in allocentric representation, and they support 172–178. the idea that to represent a new environment and build the Di Leo, J. (1970). Young children and their drawings. New York: cognitive spatial map a subject has to explore it appropri- Brunner/Mazel. Fagard, J., Rat-Fischer, L., Esseily, R., Somogyi, E., & O’Regan, J. K. ately (Mandolesi, Leggio, Spirito, & Petrosini, 2003). (2016). What does it take for an infant to learn how to use a tool by observation? Frontiers in Psychology, 1(7), 267. Acknowledgements We thank the children, parents, and Flynn, E., & Whiten, A. (2013). Dissecting children’s observational schoolteachers who made this study possible and we acknowledge the learning of complex actions through selective video displays. help by Adriano Accarino for testing some children. We are grateful Journal of Experimental Child Psychology, 116(2), 247–263. to Prof. Cecilia Guariglia and Prof. Liana Palermo for their competent Foti, F., Mazzone, L., Menghini, D., De Peppo, L., Federico, F., and kind help in analyzing and scoring the children’s drawings. This Postorino, V., et al. (2014). Learning by observation in children ´ ˆ work was supported by the Foundation Jerome Lejeune to L.M (no. with autism spectrum disorder. Psychological Medicine, 44, 886, 2011A; no. 1567, 2016B) and by PROGETTO CREME 2437–2447. ‘‘Campania Research in Experimental Medicine’’ POR Campania Foti, F., Menghini, D., Mandolesi, L., Federico, F., Vicari, S., & FSE 2007/2013 to P.B. Petrosini, L. (2013). Learning by observation: insights from Williams syndrome. PLoS One, 8(1), e53782. Compliance with ethical standards Foti, F., Menghini, D., Orlandi, E., Rufini, C., Crino ` , A., Spera, S., … Mandolesi, L. (2015). Learning by observation and learning by Conflict of interest The authors declare that they have no conflict of doing in Prader–Willi syndrome. Journal of Neurodevelopmental interest. Disorders, 7, 6. Frith, C. D., & Frith, U. (2012). Mechanisms of social cognition. Ethical standards All procedures performed in present study were in Annual Review of Psychology, 63, 287–313. accordance with the 1964 Helsinki declaration and its later amend- Gallese, V., Fadiga, L., Fogassi, L., & Rizzolatti, G. (1996). Action ments or comparable ethical standards. recognition in the premotor cortex. Brain, 119(2), 593–609. Graziano, A., Leggio, M. G., Mandolesi, L., Neri, P., Molinari, M., & Open Access This article is distributed under the terms of the Petrosini, L. (2002). Learning power of single behavioral units in Creative Commons Attribution 4.0 International License (http://crea acquisition of a complex spatial behavior: an observational tivecommons.org/licenses/by/4.0/), which permits unrestricted use, learning study in cerebellar-lesioned rats. Behavioral Neuro- distribution, and reproduction in any medium, provided you give science, 116, 116–125. appropriate credit to the original author(s) and the source, provide a Herold, K. H., & Akhtar, N. (2008). Imitative learning from a third- link to the Creative Commons license, and indicate if changes were party interaction: Relations with self-recognition and perspective made. 123 Psychological Research (2018) 82:1212–1223 1223 taking. Journal of Experimental Child Psychology, 101, psychology and cognitive neuroscience. Philosophical Transac- 114–123. tions of the Royal Society of London: Series B, Biological Iacoboni, M. (2005). Neural mechanisms of imitation. Current Sciences, 358, 491–500. Opinion Neurobiology, 15, 632–637. Meltzoff, A. N., Kuhl, P. K., Movellan, J., & Sejnowski, T. J. (2009). Keetch, K. M., Schmidt, R. A., Lee, T. D., & Young, D. E. (2005). Foundations for a new science of learning. Science, 325, Especial skills: their emergence with massive amounts of 284–288. practice. Journal of Experimental Psychology: Human Percep- Meltzoff, A. N., & Moore, M. K. (1977). Imitation of facial and tion and Performance, 31, 970–978. manual gestures by human neonates. Science, 198, 75–78. Klingberg, T. (2006). Development of a superior frontal-intraparietal Menghini, D., Vicari, S., Mandolesi, L., & Petrosini, L. (2011). Is network for visuo-spatial working memory. Neuropsychologia, learning by observation impaired in children with dyslexia? 44, 2171–2177. Neuropsychologia, 49, 1996–2003. Leggio, M. G., Molinari, M., Neri, P., Graziano, A., Mandolesi, L., & Molenberghs, P., Brander, C., Mattingley, J. B., & Cunnington, R. Petrosini, L. (2000). Representation of actions in rats: The role of (2010). The role of the superior temporal sulcus and the mirror cerebellum in learning spatial performances by observation. neuron system in imitation. Human Brain Mapping, 31(9), Proceedings of the National Academy of Science of United States 1316–1326. of America, 97, 2320–2325. Nadel, J. (2002). Imitation and imitation recognition: functional use in Lehnung, M., Leplow, B., Friege, L., Herzog, A., Ferstl, R., & preverbal infants and nonverbal children with autism. The Mehdorn, M. (1998). Development of spatial memory and Imitative Mind: Development, Evolution, and Brain Bases. In A. spatial orientation in preschoolers and primary school children. Meltzoff & W. Prinz (Eds.), The Imitative Mind (pp. 42–62). British Journal of Psychology, 89, 463–480. Cambridge: Cambridge University Press. Leplow, B., Lehnung, M., Pohl, J., Herzog, A., Ferstl, R., & Mehdorn, Nadel, J., & Butterworth, G. (1998). Imitation in infancy. Cambridge: M. (2003). Navigational place learning in children and young Cambridge University Press. adults as assessed with a standardized locomotor search task. Niederer, I., Kriemler, S., Gut, J., Hartmann, T., Schindler, C., Barral, British Journal of Psychology, 94, 299–317. J., & Puder, J. J. (2011). Relationship of aerobic fitness and Machover, K. (1949). Personality projection: In the drawing of a motor skills with memory and attention in preschoolers human figure. Springfield, IL: Charles C Thomas Publisher. (Ballabeina): A cross-sectional and longitudinal study. BioMed Mandolesi, L., Addona, F., Foti, F., Menghini, D., Petrosini, L., & Central Pediatrics, 11, 34. Vicari, S. (2009a). Spatial competences in Williams syndrome: a Overman, W. H., Pate, B. J., Moore, K., & Peuster, A. (1996). radial arm maze study. International Journal of Developmental Ontogeny of place learning in children as measured in the radial Neuroscience, 27, 205–213. arm maze, Morris search task, and open field task. Behavioral Mandolesi, L., Leggio, M. G., Spirito, F., & Petrosini, L. (2003). Neuroscience, 110, 1205–1228. Cerebellar contribution to spatial event processing: Do spatial Petrosini, L. (2007). ‘‘Do what I do’’ and ‘‘Do how I do’’: Different procedures contribute to formation of spatial declarative knowl- components of imitative learning are mediated by different edge? European Journal Neuroscience, 18, 2618–2626. neural structures. Neuroscientist, 13(4), 335–348. Mandolesi, L., Petrosini, L., Menghini, D., Addona, F., & Vicari, S. Rizzolatti, G., & Craighero, L. (2004). The mirror-neuron system. (2009b). Children’s radial arm maze performance as a function Annual Review of Neuroscience, 27, 169–192. of age and sex. International Journal of Developmental Neuro- Rizzolatti, G., Fogassi, L., & Gallese, V. (2001). Neurophysiological science, 27, 789–797. mechanisms underlying the understanding and imitation of Marshall, P. J., & Meltzoff, A. N. (2014). Neural mirroring action. Nature Reviews Neuroscience, 2(9), 661–670. mechanisms and imitation in human infants. Philosophical Rizzolatti, G., & Sinigaglia, C. (2010). The functional role of the transactions of the Royal Society of London Series B, Biological parieto-frontal mirror circuit: Interpretations and misinterpreta- sciences, 369, 2320–2325. tions. Nature Reviews Neuroscience, 11(4), 264–274. Matheson, H., Moore, C., & Akhtar, N. (2013). The development of Rohbanfard, H., & Proteau, L. (2011). Learning through observation: social learning in interactive and observational contexts. Journal A combination of expert and novice models favors learning. of Experimental Child Psychology, 114, 161–172. Experimental Brain Research, 215, 183–197. McGuigan, N. (2013). The influence of model status on the tendency Schmidt, R. A., & Bjork, R. A. (1992). New conceptualizations of of young children to over-imitate. Journal of Experimental Child practice: Common principles in three paradigms suggest new Psychology, 116, 962–969. concepts for training. Psychological Science, 3, 207–217. McGuigan, N., Gladstone, D., & Cook, L. (2012). Is the cultural Shimpi, P. M., Akhtar, N., & Moore, C. (2013). Toddlers’ imitative transmission of irrelevant tool actions in adult humans (Homo learning in interactive and observational contexts: The role of sapiens) best explained as the result of an evolved conformist age and familiarity of the model. Journal of Experimental Child bias? PLoS One, 7(12), e50863. Psychology, 116, 309–323. Meltzoff, A. N., & Andrew, N. (1995). Understanding the intentions Torriero, S., Oliveri, M., Koch, G., Caltagirone, C., & Petrosini, L. of others: Re-enactment of intended acts by 18-month-old (2007). The what and how of observational learning. Journal of children. Developmental Psychology, 31, 838–850. Cognitive Neuroscience, 19, 1656–1663. Meltzoff, A. N., & Decety, J. (2003). What imitation tells us about social cognition: A rapprochement between developmental

Journal

Psychological Research – Springer Journals

Published: Jul 20, 2017

There are no references for this article.