Introduction

Cerebral palsy (CP)—A group of neurological disorders attributed to the non-progressive disturbances of the developing fetal or infant brain [1]. Despite its non-degenerative nature, CP manifests as motor and sensory impairments, often accompanied by cognitive and communicative challenges, necessitating multidisciplinary, lifelong management. In a study conducted in 2023, CP was ranked as the third most common neurological disorder of the pediatric population in Pakistan, after behavioral disorders and epilepsy [2]. In contrast to this, the global prevalence in high-income countries is 1.6/1,000 live births [3]. This disparity may be attributed to the infrastructural differences relative to maternal and neonatal healthcare services, diagnostic tools, and rehabilitative facilities.

The etiology of CP is multifactorial, with risk factors ranging from pre-conception to antenatal and postnatal periods [4]. Such an array of origins manifests across a whole spectrum of neurological, motor, and sensory dysfunctions, including spasticity, ataxia, dyskinesia, and hypotonia, each of which is further classified topographically [5]. Spastic CP, accounting for 80% of the cases globally, frequently embodies associated gait, balance, and postural impairments [6]. Knowing the etiology is crucial for proper management and prognosis. Earlier identification of disorders, especially in the pediatric population, gives more leverage to utilize the brain’s neuroplasticity during the early neurodevelopmental window. A proactive approach towards diagnosis always helps counter the complications in the long run [7].

In general, impairments are exhibited in muscle tone, gross and fine motor functions, balance and coordination, reflexes, and posture, among others. Associative conditions include oromotor dysfunction, visual and hearing difficulties, sensory impairments, and seizures [8]. Such deficits increase fall risk, hinder proper gait and ability to perform activities of daily living (ADLs), and lead to decreased functional performance and quality of life (QoL) [9]. Mobility limitation in CP is visible as difficulty in walking, transitioning between positions such as sit to stand, and maintaining correct static and/or dynamic posture. Children with CP often experience even simple daily tasks as physically demanding and difficult [10, 11].

CP management involves a multidisciplinary team ranging from neurologists to physical therapists, speech therapists, and nutritionists. Early diagnosis and intervention are crucial to utilize the neuroplasticity of the infant brain and early childhood to maximize the functional gains [12]. Previous research has brought attention to the utility of action observation therapy (AOT) across a range of neurological populations, including post-stroke patients and patients with Parkinson’s disorder. However, its clinical integration into pediatric neurorehabilitation, exclusively for children with CP, remains underexplored. AOT is a relatively novel and emerging approach rooted in the principles of the mirror neuron system. First described in the early 1990s, researchers have now reached a basic agreement regarding the involvement of mirror neurons, located in different brain regions including the premotor and parietal cortices, in the perception of motor actions, human speech discrimination, and imitative responses [13]. AOT involves structured observation of goal-directed actions followed by their execution and repetition. This sequence stimulates neural circuits involved in motor planning and imitation, facilitating motor learning even in the absence of voluntary initiation.

How mirror neurons work can be understood through the action of yawning. When you see someone yawn, the mirror neurons in your brain get activated and start interpreting the motor task behind the action, and you feel the urge to yawn. This is termed the motor system’s ability to “reflect” observed actions and recruit the same neural pathways to produce a similar response [14]. The same neural firing occurs during the mere observation as when performing it, enabling the motor system to become more responsive and receptive over time.

A typical AOT session consists of the following phases [15]:

• 1.

  Observation phase: The patient watches short clips of the task to be performed (e.g., picking up a cup). These actions are intended to engage the mirror neurons.

• 2.

  Execution phase: The patient is asked to perform the task as shown before. This step is for reinforcing motor learning.

• 3.

  Repetition phase: The steps are repeated several times to promote motor memory.

The sessions are 15–30 minutes long, with a frequency of 3–5 times a week for several weeks, depending on the protocol [16].

The heterogeneity in delivery modes for AOT, spanning from therapist-led, parent-delivered to video-based, and virtual reality-enhanced formats, provides flexibility but also poses obstacles to widespread implementation and standardization [17–20]. As for the CP populations to be specific, AOT has been investigated both as a stand-alone treatment and in conjunction with other strategies such as Constraint Induced Movement Therapy (CIMT) or RP (repeated practice) for upper and lower limb impairments, with the majority of the studies focusing on the upper limb [20–22]. Studies that targeted the lower limb have reported significant improvements with the use of AOT [23, 24].

Given these considerations, this study aims to investigate the effectiveness of AOT in improving functional performance, specifically balance and mobility, in children with spastic CP. By contributing empirical evidence to the field of pediatric neurorehabilitation, this research seeks to evaluate AOT’s potential to enhance functional independence and inform future clinical protocols.

Materials and methods

Results

Demographics

From September 16th, 2024, to March 25th, 2025, out of the total 43 children screened for eligibility, 28 met the inclusion criteria and were randomly allocated to either the experimental (n = 14) or control (n = 14) groups. The intervention protocol was successfully delivered in both groups. There were no drop-outs during the study (see Figure 1).

Display full size

Figure 1. CONSORT diagram showing the data collection procedure.

The baseline demographic information of the enrolled participants has been displayed in the table below. The key demographic variables include the prevalence of age, gender, spastic CP subtype, and GMFCS levels among the participants (see Table 2).

Table 2. Key demographic data for both groups.

Parameters	AOT group	Control group
Age in years (mean ± SD)	4.71 ± 2.2	4.43 ± 1.83
Gender, male/female (n, %)	6 (42.9)/8 (57.1)	6 (42.9)/8 (57.1)
Spastic CP subtype (n, %)
Spastic monoplegia	2 (14.3%)	4 (28.6%)
Spastic diplegic	9 (64.3%)	7 (50.0%)
Spastic hemiplegic	3 (21.4%)	3 (21.4%)
GMFCS level (I/II/III)	3/8/3	6/6/2

Display full size

AOT: action observation therapy; CP: cerebral palsy; GMFCS: Gross Motor Function Classification System.

A similar gender distribution across both study groups, showing a similar male (43%) to female (57%) ratio in both arms. In the distribution of spastic CP subtypes across both study groups, monoplegia spastic CP was the most prevalent subtype in the experimental group, while diplegic spastic CP was more prevalent in the control group. For GMFCS levels, the majority of the children were GMFCS level II in the experimental group, and an equal number of children were GMFCS level I and II in the control group.

Discussion

This study primarily aimed at evaluating the effects of AOT on functional performance in children with CP, gauged under the two domains of balance and mobility. While AOT and its potential efficacy on different neurological disorders have gained increased attention, evidence of its effectiveness in CP remains contradictory. In the present investigation, between-group analysis did not reveal a statistically significant difference; both groups demonstrated within-group improvements after intervention. These findings supported the notion that both therapeutic approaches were beneficial for the children.

All 28 children recruited in the study showed improvement in functional performance in both domains of balance and mobility, but the difference between the therapeutic approaches was not statistically significant enough to establish superiority. The complete adherence and no dropouts were likely facilitated by the play-based format of the exercise protocol. Continued guidance, instructions, and assistance to the children throughout the protocol ensured they understood the game tasks. It is also critical to acknowledge that the children’s emotions and mood states occasionally affected how involved they were in treatment sessions. These incidents were rare and had little impact on the overall intervention procedure, but they did occasionally provide slight difficulties in keeping the focus and flow of the session.

The findings in the current study are consistent with previous studies and indicate that the benefits of AOT may be on par with the conventional approaches, but the superiority remains inconclusive. Existing literature has reported variable results, which may offer some context for better understanding our results. Consistent with our findings, Kirkpatrick et al. [20] have conducted a randomized controlled trial on parent-delivered AOT in children with unilateral CP. The researchers discovered no statistically significant difference between RP alone and AOT paired with RP. Despite the differences in the modes of delivery—home-based versus physiotherapist-led—both techniques produced improvements. This implies that the repetition of functional tasks itself may be the primary contributor to reported benefits.

A 2020 study found that combining AOT with Whole-Body Vibration (WBV) for children with spastic CP improved gross motor function and gait measurements [22]. The combined intervention group outperformed the control group, implying that AOT can be more effective when combined with other techniques. In contrast, the current study assessed AOT as a stand-alone intervention, which may have limited its overall effectiveness. A 2019 meta-analysis on AOT in individuals with brain damage demonstrated significant improvement in upper and lower limb function, particularly upper limb activity (p < 0.001) [30]. However, the substantial heterogeneity in sample characteristics and intervention techniques created a significant hurdle to standardizing AOT protocols. The authors emphasized the relevance of personalized and condition-specific AOT applications, which aligns with our findings, especially considering our focus on balance and mobility in a heterogeneous group of children with CP.

Other studies have reported favorable outcomes with AOT. Jeong and Lee [23] discovered that AOT greatly enhanced gross motor function and reduced ankle stiffness in children with spastic CP. Their more intensive protocol (five sessions per week of one hour each) may explain the better results compared to our shorter, less frequent sessions. This discrepancy shows that intervention intensity is critical to maximizing the effects of AOT. Having a similar stance regarding the importance of intensity, Simon-Martinez et al. [21] investigated the effects of AOT paired with CIMT throughout a 9-day intensive training period. Although the between-group analysis was not statistically significant, the AOT group displayed a slightly higher post-intervention score (p = 0.02).

A recent study in 2024 recruited spastic diplegic CP and found significant gains in lower limb Gross Motor Function using an AOT-based protocol [24]. Although the intervention design was similar to ours, differences in participant characteristics and outcome measures are likely to have influenced the findings. Their emphasis on lower-limb mobility using a subscale of the GMFM contrasts with our more comprehensive assessment of balance and mobility using the GMFM-88 and PBS. This broader scope may have resulted in more nuanced but less noticeable gains in our sample. It is important to mention that AOT may influence neuroplasticity in ways that behavioral or functional assessments alone may not fully account for. Post-MRI scans reported in a 2024 review evaluating the efficacy of AOT in upper limb function in CP children revealed increased activation of the frontal and parietal areas of the brain when involved in hand-object interaction tasks (p < 0.001), which supported the notion of AOT enhancing neural engagement [24].

To summarize, AOT holds promise as an intervention for CP children, but its effectiveness highly depends on factors such as intervention intensity, combination with other treatments, and population-specific modifications. It is especially hard to yield good results with a “one size fits all” protocol, particularly in the pediatric population. Several limitations of this study should be acknowledged. First, the inclusion of participants across various types of spastic CP and a broad range of GMFCS levels may have produced heterogeneity in responsiveness to the intervention, potentially influencing the uniformity of the results. Secondly, as discussed earlier, a few participants exhibited a limited attention span, leading to lengthened sessions and minor difficulty in adherence to the protocol, as expected in managing the pediatric population. Due to the nature of the intervention, it was only right to let the administrator of the intervention be the assessor; however, to minimize observer bias, it was prioritized to use the validated tools only, i.e., PBS and GMFM-88. It is of equal importance to acknowledge that the sample size was small due to the involvement of a specific subgroup. The smaller sample facilitated intensive care and accurate data collection as it involved the pediatric population. However, such results provide the foundation for large-scale research in the future.

Future research should focus on refining AOT techniques, determining optimal dosage and duration, and using neuroimaging or other objective tools to better understand its impact. Standardizing these characteristics will be critical to establishing AOT as a reliable, evidence-based therapy in pediatric neurorehabilitation.