Grapho-motor imitation training in children with handwriting difficulties: A single-center pilot study

Abstract

Imitation is a crucial process for learning and brain development. It is based on the mirror neuron mechanism and underlies our understanding of actions and the gestures of others. Some researchers hypothesized a possible correlation between a low functioning mirror neuron system (MNS) and developmental coordination disorder, including dysgraphic deficit. However, no studies have verified whether imitation of graphic gestures by exploiting the properties of the MNS could improve handwriting. This study evaluates the effects of imitation training of handwriting in five children with handwriting difficulties aged 8 to 10 years. The training lasted for five months, and was undertaken three times a week, for a total number of 60 sessions for each child. Before and after the training, we evaluated the degree of handwriting impairment using the Concise Evaluation Scale for Children’s Handwriting (BHK). Our results suggest that handwriting imitation training produced a significant qualitative change in the children’s writing, likely due to exercises that stimulated fine motor imitation. Furthermore, the imitation also involved ergonomic and biomechanical aspects relevant to improving imitative writing after observing the model. Each child has therefore reached an adequate level of writing, suggesting the effectiveness of the proposed intervention.

Keywords:

• handwriting difficulty
• mirror neuron system
• imitation
• action goal
• gestures

1. Introduction

Imitation is the capacity to reproduce novel movements performed by others and is fundamental for cognition and development (Subiaul et al., 2016). The most important aspect of this capacity is imitation learning (Del Giudice et al., 2009), which involves transforming a new observed action in an executed action identical or similar to the observed one (Buccino et al., 2004).

According to Vygotskij (1960), imitation plays a central role in learning and development. By acting and behaving in a particular way and providing instructions on how to act, the adult individual stimulates the child to imitate and thereby develop new skills. From birth, learning and maturation act reciprocally (Vygotskij, 1978), and according to Vygotskij, imitation constitutes intelligent behavior (Vygotskij, 1960).

Studies on neonatal imitation have suggested that imitation is an innate mechanism. Meltzoff and Moore (1977; 1983; 1997) observed that newborns imitate some communicative facial expressions performed by adults, even within the first hours of life. Neonatal imitation, which does not involve conscious understanding, is a mirroring behavior, whose purpose is most likely to strengthen the attachment between the newborn and the caregiver. This type of imitation lasts for a few months, which is the time necessary to reinforce bonding. Intentional imitation only occurs after the eighth month.

The discovery of mirror neurons has recently provided a neuroscientific explanation for imitative behavior. These neurons, which were originally discovered in the monkey ventral premotor cortex (DiPellegrino et al., 1992; Gallese et al., 1996; Rizzolatti et al., 1996) and then reported in the inferior parietal cortex (Fogassi, Lupino, 2005; Rozzi et al., 2008) are activated when a monkey executes a goal-directed motor act and when it observes another individual, either a monkey or human, performing the same or a very similar motor act. Thus, these neurons engender both a motor and a visual response.

A mirror neuron system (MNS) has also been discovered in humans, through the use of electrophysiological and neuroimaging techniques. In humans, observing and executing motor acts activates a circuit that includes the temporal cortex, the inferior parietal lobule (IPL), the ventral premotor cortex and the posterior part of the inferior frontal gyrus (IFG) (Caspers et al., 2010; Cattaneo & Rizzolatti, 2009; Molenberghs et al., 2012; Rizzolatti et al., 2014). It is known that both IPL and IFG activate for goal related hand motor acts, such as object grasping and manipulation (Binkofski et al., 1999; Culham et al., 2003; Errante et al., 2021). The hand motor representations contained in these areas activate also when the same motor acts are simply observed (Cattaneo & Rizzolatti, 2009). It has been proposed that this mechanism, that matches observation and execution of the same motor acts, allows observers to automatically understand others’ actions (Rizzolatti et al., 2001). The MNS plays an important role in imitation in humans. For example, after observing and imitating finger movements both the IFG and IPL were activated (Iacoboni, 1999). Buccino et al. (2004) showed that the MNS was activated during imitation learning, such as in naïve subjects copying guitar chords shown by an expert guitarist, using functional magnetic resonance imaging (FMRI). Activation of the MNS was even higher during active imitation by participants inside the scanner. This increased activation can be explained by the mirror neuron mechanism allowing the participant to recognize the single acts that compose the chord. Next, these acts are reconstructed in sequence according to the model through the collaboration of the MNS with the prefrontal cortex, which is known for its role in working memory and sequence building.

The strong involvement of the MNS in imitation learning has led to its potential use in the rehabilitation of patients with motor impairments, provided their MNS system remains intact. Studies based on action observation followed by its reproduction showed that this method produced motor improvements in stroke patients, children with cerebral palsy, and in patients with Parkinson’s disease (Abbruzzese et al., 2015; Ertelt et al., 2007; Sgandurra et al., 2013).

Recently, a possible correlation between a low functioning MNS and developmental coordination disorder (DCD) has been suggested (Licari et al., 2015; Reynolds et al., 2015, 2017; Werner et al., 2012). Individuals with DCD have difficulty with the fine motor execution of writing (APA, 2013; Alstad et al., 2015; Rosenblum & Engel-Yeger, 2014) and suffer from clumsiness in their motor abilities, such as dressing, tying shoelaces, or sports (APA, 2013). In these individuals, the level of impairment relates to the low or anomalous activation of the MNS.

Werner et al. (2012) analyzed the existing literature on neuroimaging studies of DCD participants and hypothesized that the MNS could be malfunctioning, resulting in motor and imitation deficits. More recently, Reynolds et al. (2017) correlated the imitation deficit with the abnormal functioning of the MNS. According to Reynolds (2017), the deficit in imitation shown by some children with a presumptive diagnosis of DCD is responsible for their difficulty in motor learning, which prevents them from forming new and complex postures and gesture sequences. Furthermore, these children showed slower responses than non-DCD control subjects.

Few studies have investigated dysgraphia (part of DCD) and its treatment. This writing impairment can be comorbid with DCD, but it can also be present within a specific learning disorder (SLD), although it is not clearly defined by the Diagnostic and Statistical Manual of Mental Disorders (DSM–5) (APA, 2013). Writing difficulty is only mentioned within SLD as an impairment of written expression (Rosenblum, 2018). The difficulty in the graphic components of handwriting is often defined with the word “dysgraphia” (Döhla & Heim, 2016; Smits-Engelsman & Schoemaker, 2017).

Other terms used in the literature are “poor writen” (Nijhuis van der Sanden & Overvelde, 2013), “poor handwriting” (Feder & Majnemer, 2007) or “handwriting difficulties” (Rosenblum et al., 2003) (Guidelines of Istituto Superiore di Sanità, (2022)).

Hamstra-Bletz and Blote (1993) had already clarified how the handwriting disorder is characterized by the difficulty in the mechanical realization of written language, especially italics. Children with this impairment have normal intelligence and they do not show neurological and perceptual-motor problems (Engel-Yeger & Rosenblum, 2010).

The difficulty in handwriting shown by dysgraphic children can be related to some abnormality in the motor representation of the complex gesture required for producing a written word, that involves both proximal (arm and forearm) and distal (wrist and fingers) components. Very likely this neural representation is contained in parietal and premotor cortex, and, as many other types of representation, could be sensitive to learning-induced plastic changes. Learning could be simply based on a trial-and-error physical exercise, or it could be triggered by observation of a model, as it occurs in imitation. Since, as we mentioned above, this letter constitutes a faster and more efficient learning strategy, it seems feasible to try to apply it to dysgraphic children. The rationale is that observation of a healthy model (an adult individual) performing a normal writing behavior should automatically retrieve, thanks to the mirror neuron mechanism, the corresponding motor representations in the parietal and premotor cortex of the observer. The following reproduction, by the observing child, of the gesture done by the observed model should induce a change in these representations, through a process of gradual approximation.

Thus, with this preliminary study we aim at verifying whether children who observe the experimenter writing can imitate her/his motor gestures and improve the quality of their writing. To address this issue, we selected five children with different profiles of writing difficulties and evaluated the effectiveness of the imitative handwriting training created by us. The tool we used for this evaluation is the BHK (DiBrina & Rossini, 2011; E. Hamstra-Bletz et al., 1987). We chose this tool because it offers high sensitivity in identifying children with various types of writing difficulties and is also used at the international level.

2. Methods

2.1. Participants

Five Italian children (four boys, one girl, all right-handed) with handwriting difficulties participated in the study. The children were recruited from local schools in the middle of Italy. The inclusion criteria were: (1) they had not previously received any type of grapho-motor rehabilitation treatment; (2) they had never written in italics, and they were not certified as being disabled (IQ ≥ 70); (3) all the participants’ parents signed the informed consent/assent after a study briefing according to the Psychologists Ethical Principles and the Code of Conduct of the American Psychological Association (2010). In addition, the International Committee of Medical Journal Editors (ICMJE) requirements on privacy and informed consent of study participants were met.

This study was conducted according to the guidelines reported in the Declaration of Helsinki.

The four boys received a diagnosis of handwriting difficulty from the Public Service of Neuropsychiatry and Rehabilitation for their developmental age. All diagnoses were conducted according to the international parameters of the DSM-5 (APA, 2013). While the girl did not have an official diagnosis of handwriting difficulty, she presented with clear difficulties writing during the test. Her test score was borderline normal. Even at school, in fact, the girl had failed to automate italics writing.

The anamnesis of each child did not show any delays in language and neuromotor development and no neurological disorders.

Table 1 lists the demographic characteristics and diagnoses for each child.

Table 1. Demographic characteristics and diagnoses

Cases	Gender	Age (year)	PS class	Writing style	IQ	Learning disorders
AG	boy	8.10	IV	block letters	114	dyslexia, dysorthography, dysgraphia
MS	boy	10.01	V	block letters	109	dyslexia, dysorthography, dysgraphia
ME	boy	10.03	V	block letters	106	dysorthography, dysgraphia
GC	girl	8.09	IV	block letters	106	**
AP	boy	9.04	IV	block letters	80	dyslexia, dysorthography, dysgraphia, DCD

PS = Primary School; IQ = Intelligence Quotient; DCD = Developmental Coordination Disorder

(**) The child has many problems with reading and writing. No diagnosis on grapho-motor aspects was reported, although teachers claim she has dysgraphia.

2.2. Training

The children participated in three sessions per week of imitation training for 20 consecutive weeks and took part in a total of 60 sessions. Each session lasted 15 minutes. The material used consisted of a series of matrices created on lined sheets of notebooks that children normally used in the second class of primary school. The experimenters had previously written fables and stories in clear handwriting on these sheets using medium-sized italic letters.

The experimenter and child had identical matrices in front of them, and the child had to trace the matrix after observing the experimenter performing the same actions. Specifically, the experimenter invited the child to carefully observe how he held the pen and the motor movements he performed to draw the underlying line. At this point, the experimenter traced a line of text and then stopped. Then the child was instructed to do the same in the same way. This procedure was repeated for six lines of the matrix.

During each session, the experimenter (right-handed) sat behind a desk to the right of the child so that the child could easily watch all the movements made by the experimenter from an egocentric perspective. The child was invited to imitate the posture of the experimenter in terms of how to sit and how to position their arm and hand. In particular, the experimenter instructed the child to observe the position of his left hand to hold the sheet and stop it from moving. Additionally, those children who tended to move their right hand during the observation phase (for example, drumming their fingers on the table) were asked to keep their hands still. The experimenter and the child used the same type of pen, which was carefully chosen for its smoothness and comfortable grip, to ensure the stroke was a medium size.

At the end of each session, the experimenter used observation matrices to record the ergonomic and postural aspects worked on during the sessions. This provided useful information on changes in the children’s biomechanical, ergonomic and postural changes occurring during the training.

To evaluate the training effects, each child was tested on their grapho-motor abilities two days before training (T0) and two days after the end of training (T1).

2.3. Measures

2.3.1. The concise evaluation scale for children’s handwriting (BHK)

The BHK (DiBrina & Rossini, 2011; Hamstra-Blez et al., 1987) is an evaluation scale used in basic research and clinical practice. It is a standardized test in multiple languages (Netherlands, Switzerland, Germany, France, Italy). This tool “is used in the neuropsychological evaluation of specific developmental disorders and more generally in the detection of a bad use of fine motor skills, even in the absence of frank neuropsychological pathologies” (Di Brina, Rossi, p. 7–8). It is an analytical scale based on the assumption that there is a relationship between the general aspect of the handwriting and some predefined criteria. The text written during the test is evaluated for the parameter of readability, i.e. the quality of writing, through 13 indices, grouped into 3 clusters: organization of written words, formation of letters, fine motor skills. More specifically, the child has to copy a standard text onto a blank sheet and the items included in the evaluation are: writing size, unaligned left border, swinging progress of writing line, insufficient space between words, sharp angles or elongated links, links interrupted between letters, letter collision, letters of irregular size, inconsistent distance between letters with or without extension, atypical letters, ambiguous letter shape, readjusted or retraced letters, and unstable graphic stroke. The sum of the scores obtained in all these items give an overall score allowing to detect dysgraphia in children. The overall (raw) score makes it possible to calculate the standard deviation (SD), by subtracting the total BHK score of each subject from the normative sample average (20.3 for boys, 18.1 for girls) and then dividing the obtained result by the SD of the normative sample (6.1 for boys, 6.7 for girls).

Furthermore, we calculated the 95% confidence interval (CI) of the means of the normative samples separately for boys (CI 95% 19.6–21) and girls (CI 95% 16–18.88) to compare the scores achieved by each child involved in the present study with the reference sample both at T0 and T1.

2.3.2. Observation of biomechanical and postural ergonomics aspects

Observation is considered functional because it acts as a research methodology and a means to collect data (Lis & Venuti, 1996; Pearsall, 1970; Williamson, 2000). Observation is used as a systematic recording method for events, behaviors, and artefacts in the specific setting under study (Marshall & Rossman, 1999). According to Glazier (1985), observation is a functional tool for qualitative evaluation, but it requires the formation of predetermined categories to record the various types of activity or behavior. Therefore, observation is used to gather information on possible variations in the motor skills used during handwriting using morphological and functional categories (Blurton Jones, 1972; McGrew, 1972). The use of morphological and functional categories can be used to detect imitation of such behavior.

Thus, observations are systematic and are based on a model of fine motor behavior.

Rosenblum et al. (2006) noted the scarcity of studies on the relationship between handwriting and biomechanical ergonomic factors useful to improve handwriting. In this study, we considered international works on the relationship between handwriting and biomechanical ergonomic factors to improve handwriting, such as those of Schneck and Henderson, (1990), Feder and Majnemer (2007), Rosemblum et al. (2006), and Smith-Zuzovsky and Exner (2004). We based our observation categories on the existing literature and direct observations of participants’ behavior. Table 2 lists the categories and the parameters considered for the systematic observations.

Table 2. Categories and observation parameters

CATEGORIES	OBSERVATION PARAMETERS
Body posture during handwriting	- feet aligned and placed on the floor - hips and lower back leaning on the back of the chair - knees bent at 90°
Arm and wrist position	- flexed elbows - forearm leaning on the table - wrist sufficiently soft and flexible
Type of pen grip	- dynamic tripod grip - lateral tripod grip
Trace strength	- trace not too marked (it must not be visible on the back of the sheet)
Direction of letters	- anticlockwise direction

3. Results

Table 3 lists the raw scores and SDs obtained using the BHK scale for each participant at T0 and T1. The scores at T0 for all participants are higher than the upper limit of the CI calculated using the means of the raw scores of the normative sample, subdivided by gender. The SDs for all the boys are < −2, suggesting they all have moderate to high difficulty in handwriting. AP had the highest level of difficulty, while GC presented moderate difficulty.

Table 3. Raw score and standard deviation (SD) of the pre- and post-test scores of each participant

	T0		T1		ΔSD	95% CI Raw score of the normative sample
	Raw score	SD	Raw score	SD
AG	49	−4.70	17	0.54	5.24	19.6–21
MS	45	−4.04	18	0.37	4.41	19.6–21
ME	35	−2.40	20	0.04	2.44	19.6–21
GC	29	−1.62	16	0.31	1.93	16–18.88
AP	49	−4.70	25	−0.77	3.93	19.6–21

At T1, all participants other than AP showed a raw score below the upper limit of the CI and were within the normative curve. Additionally, all the SDs were close to 0, indicating that all the participants showed an improved handwriting ability.

SD = Standard deviation; ΔSD = Change in Standard deviation; CI = Confidence interval; T0 = Before the start of training; T1 = After completion of training

Figures 1 and 2 are examples of writing in two children (AP and MS, respectively) in the pre-test (T0) and in the post-test (T1) after a 60 sessions training.

Figure 1. Example of the BHK test performed by AP at T0 and T1.

Figure 2. Example of the BHK test performed by MS at T0 and T1.

Figure 3 shows two examples of free writing. Note that even after just a few imitative training sessions, the writing has become more readable.

Figure 3. Example of free writing performed by AP and MS after 19 sessions.

3.1. Observation summaries

We analyzed the observations conducted during each treatment training session to show the progressive improvement of the motor parameters. The writing improvement seen in all five children (as demonstrated by the data shown in Table 3) is most likely due to them imitating the experimenter throughout the grapho-motor training and imitating the postural and ergonomic aspects. All the aspects reported in the observations are significant because they impact the grapho-motor aspects.

Observation analyses of the first indicator (body posture during handwriting) in the initial sessions revealed that not all the children positioned themselves correctly in front of the writing table; AG and MS were not seated correctly. Both held their bodies rotated to the left (lateral position), with the right arm held above the tabletop. Within the first ten sessions, both children showed changes in this specific indicator; MS mimicked the position of the experimenter’s body by the third session and AG by the tenth session.

In terms of the second indicator (position of the arm and wrist), at the start, the right arms of all the subjects were extended on the tabletop instead of being in a flexed position. Also, all the children held their arms very stiffly, making it difficult to position them comfortably for writing. For example, it took a great deal of effort for AP to hold the pen in his hand. His arm was much more rigid than the other boys, even when the position was the same. Within the first ten sessions, four out of five children, including AP, could change their arm positions gradually. More specifically, GC (after the third session), AG and AP (after the first week of sessions), and ME (after the ninth session) were able to bend their elbows and position their writing arms more towards the axis of their bodies. This allowed greater mobility, avoided them having to extend the arm above the table, and allowed greater fluidity in their writing. However, MS could not imitate entirely the position of the arm, which he continued to extend over the writing table, even after all 60 sessions.

Observation analysis of the position of the right hand and wrist showed that the wrists of AG, MS, ME, and GC were initially very rigid. ME, GC, and AP kept his wrist and hand bent inwards, especially at the end of the writing line. AP’s wrist appeared to be almost paralyzed and was stiffer than the other boys. However, over time, some changes in the movement of the wrist and hand became apparent. Most children (four out of five) showed changes in the wrist and hand movements that appeared softer and more dynamic (flexible) within the first 30 sessions. GC showed improvement from the 3^rd session, ME from the 15^th, AP from the seventh, and AG after the 30^th. However, MS retained the constant stiffness in the arm and wrist, even after 60 training sessions.

Observations of the third indicator (type of pen grip), revealed that two out of five boys had an immature grip (Schneck & Henderson, 1990). AG and AP had similar grips; they held the thumb over the index finger, which was kept fully extended (cross thumb grip) (Schneck & Henderson, 1990). AP also held the pen very near to the tip, preventing him from seeing what he was writing. From the 30^th session, AG gradually assumed a side tripod posture which transformed into a dynamic tripod grip (a more developed position) by the end of the training course (Schneck & Henderson, 1990, Feder, Maynemer, 2007). In AP, the cross thumb grip remained unchanged after all 60 training sessions.

At the start, three out of five boys had a marked and not very smooth type of graphic trace, which was particularly heavy in AG and AP. AGs graphic trace was visible on the back of the sheet and on the one underneath, while APs graphic line was so heavy that it pierced the writing sheet—probably because he sometimes went back over and retraced his letters. MS never detached his pen from the sheet while writing.

All three boys began to demonstrate changes in their graphic traces after half of the training sessions. AGs writing line became less pronounced, and AP no longer retraced his letters. MS required more input from the experimenter, but he eventually succeeded in copying the experimenter by session 30 and gradually showed improved spontaneity in the movements that he learned to imitate.

At the start, all the boys wrote letters in the wrong direction. For example, they wrote “a” and “o” in a clockwise direction instead of counterclockwise, and “t” and “p” from the bottom upwards. After the training, all the students managed to imitate correct letter construction; GC after the 2^nd session, ME after the 3^rd session, MS after the 7^th session, AG after the 8^th session, and AP after the 9^th training session.

In conclusion, this pilot study allowed us to identify macro- and micro- analytical observation criteria to construct grids that can guide during individual sessions. Further studies involving the use of video recordings of the individual sessions are needed to make the grids more relevant and to be able to measure them quantitatively.

4. Discussion

The capacity to imitate is crucial for learning and development (Vygotskij, 1960). It implies learning a new motor model or sequence (Buccino et al., 2004). This study evaluated whether the use of imitation training for handwriting was effective. The subjects included five children aged 8 to 10 years; four (boys) with dysgraphia and one (girl) with delayed handwriting who had not previously received interventional therapy. We ascertained whether the subjects could functionally imitate the experimenter’s gestures to write in italics, legibly, and in a style similar to that of the used matrices.

The results of the BHK test administered at T0 and T1 show that the training was successful as all the children had scores within the normative curve at the end of the training, indicating correct and readable writing. In particular, AG and MS demonstrate better performance at T1 than 95% of the normative sample. Overall, there was a noticeable improvement in the quality of handwriting in all participants.

This demonstrates how imitation learning enabled the children to learn and incorporate functional motor schemas. Our observations also showed that ergonomic and biomechanical aspects were imitated after a few sessions of observation of the experimenter’s motor model. Consequently, all the children achieved appropriate and functional ergonomic and biomechanical levels in their writing. One child diagnosed as having DCD with severe and persistent symptoms (AG) was able to compensate for the relevant grapho-motor, postural, and motor deficits due to the imitation training.

Our results agree with the existing literature on observation-based imitation, both in healthy participants and patients (Buccino, 2004). In particular, children and adults with motor impairments benefit from action observation-based rehabilitation (Abruzzese et al., 2015; Ertelt et al., 2007; Franceschini et al., 2012; Sgandurra et al., 2013). The improvements observed following therapy are likely due to the MNS exploiting two main components. First, as shown previously, the MNS codes for the goal of the observed actions (Buccino et al., 2001; Gazzolla et al., 2007; Gazzolla et al., 2007; Hamilton & Grafton, 2008), thus, it is likely that the student strives for the same goal as the observed action. Second, actions are characterized by kinematic parameters, so students may also take advantage from the decoding of this information. Within the neural network activated during action observation, two primary parieto-premotor circuits exist; a ventral circuit and a dorsal circuit which appears to decode the kinematic aspects of observed actions (Casile et al., 2010; Errante & Fogassi, 2019). In our study, the BHK scores and the ergonomic-biomechanical results suggest that the participants attempted to imitate the task objective and some kinematic aspects, including the posture needed to improve their writing. Observations aimed at imitating meaningful gestures activate the dorsal part of the action observation network (Grèzes et al., 1998). Despite most previous studies focusing on goal-related actions, the accumulated evidence clarifies the underlying neural mechanisms for subjects with grapho-motor impairments to improve their writing through model observations.

We hypothesize that the successful imitative training we conducted for children with dysgraphia could positively affect other learning disorders. It has been suggested that problems in grapho-motor abilities are linked to dyslexia and dysorthography (Gimenez et al., 2014). Furthermore, good grapho-motor skills could have direct effects on children learning to read. This is suggested because the motor experience of creating letters with a pen can help children in discriminating their main features. This leads to the construction of a more accurate brain representation that, in turn, facilitates letter recognition, and consequently, more fluent reading (Gimenez et al., 2014). These changes do not occur when people learn to write by typing on a computer keyboard (Longcamp et al., 2003; Longcamp et al., 2005a; Longcamp et al., 2005b). Furthermore, as shown by James and Engelhardt (2012), in preschool children, brain activation is higher in the cortical regions related to reading (e.g., the fusiform gyrus, posterior parietal cortex, and inferior frontal gyrus), when handwriting was compared with typing.

Finally, Gosse and Van Reybroeck (2020) showed that handwriting and orthography functionally interact during writing. This interaction can depend on the reciprocal influence between the grapho-motor and orthographic capacities of writing.

5. Conclusions and limitations of this study

Our results suggest that handwriting imitation training produced a significant qualitative change in the children’s writing, likely due to exercises that stimulated fine motor imitation. Furthermore, the imitation also involved ergonomic and biomechanical aspects relevant to improving imitative writing after observing the model. In summary, our results suggest that further developing this approach could impact clinical, rehabilitation, and didactic/pedagogical fields of study.

Today, even though the creation of text does not require handwriting, it remains one of the most used forms of writing in school, and some studies have shown that teaching handwriting can help dyslexic children improve their reading ability (Hebert et al., 2018).

The sample included in this study was limited; however, it must be borne in mind that the percentage of school-age children with dysgraphia is also low. Furthermore, our sample selection did consider some fundamental criteria, such as: a diagnosis of dysgraphia, age between 8 and 10 years, no previous history of rehabilitation treatment, absence of italics writing capacity, and a normal IQ. In the future, we aim to perform imitation training using a larger sample and, if possible, to conduct a parallel neuroimaging study.

Acknowledgments

The authors thank all the children and their parents who collaborated in this study and made this research possible. The authors would also like to thank the “Words that Fly” (A.PA.V) Association, which is responsible for providing support to children with specific learning disorders and their families, for its organizational support.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This research received no external funding.