Introduction: Epilepsy is one of the most common neurological diseases in childhood and can adversely affect various aspects of a child's functioning. It is defined as a disorder in brain function characterized by recurrent seizures caused by excessive activity of brain cells, leading to disturbances in motor function, sensation, consciousness, behavior, and emotions. According to the guidelines of the International League Against Epilepsy from 2017, the classification of epilepsy is based on the seizure type, epilepsy type, and etiology of the disease, as well as establishing an epilepsy syndrome if possible. The illness is heterogeneous in terms of various manifestations of epileptic seizures, causes, and its impact on a person's life. Factors related to the disease itself, treatment demands, limitations in daily functioning, and psychosocial factors contribute to its complexity. The most highlighted comorbidities include ADHD, learning and memory difficulties, intellectual deficits, anxiety and depression, behavioral disorders, and challenges in social relations. A major source of concern for parents is the increased risk for developmental and cognitive difficulties and it is justified by the fact that children with epilepsy have a tenfold higher risk of cognitive deficits compared to healthy peers and that cognitive difficulties are more frequent in children with epilepsy than in those suffering from other chronic illnesses. The relationship between cognitive functioning and epilepsy is complex, with highly heterogeneous cognitive performance in children with epilepsy where degree of cognitive impairment varies greatly, ranging from very mild to severe and long-term progressive forms. Studies addressing this issue mostly confirm that children with epilepsy as a group achieve scores within the average range on measures of global development and general intelligence, but shifted towards lower values. Also, if we want to understand their cognitive functioning even better, that is possible by observing their efficasy in various developmental and cognitive areas. The research on developmental profiles of preschool children with epilepsy is not entirely consistent, but majority tends to indicate lower (still average) performance on visuomotor tasks involving block manipulation, figure arrangement, and visuomotor integration. Although, verbal intelligence quotient (IQ) scores typically do not significantly differ between children with epilepsy and control groups, this does not necessarily indicate unimpaired language abilities. A specific and detailed language assessment conducted by speech therapists can highlight issues in this developmental area, too. When discussing profile of cognitive functions of school-aged children with epilepsy, certain areas as working memory and processing speed, commonly show lower scores. One of the most important objectives of psychological assessment of preschool children (including those with epilepsy) is predicting future cognitive functioning on the developmental assessement results. Given the lack of studies on the stability and the predictive validity of developmental assessment results in children with epilepsy, we focused on the insights of research of extremely low birth weight and premature born children. Those studies highlight that testing developmental milestones after age 2, time interval between the initial assessment and follow-up, and the instruments used, are all important factors for the stability and the predictive value of developmental assessment. In the Republic of Croatia, the most commonly used developmental scale is the Čuturić Developmental Test (RTČ-P), which has satisfactory psychometric characteristics, is practical for use, and provides a framework for assessing progress in various developmental areas. Research on the stability of this scale in clinical samples of children is very rare, and in more specific clinical samples (as a group of patients with epilepsy) does not exist. Such data are crucial in clinical practice with patients and their parents (including children with epilepsy and their families) to make the diagnostic process more precise and to provide targeted recommendations aimed at enhancing a child's strengths and mitigating weaknesses. That is why neuropsychological assessment has played an important role in the clinical monitoring of epilepsy for over fifty years. Methodology: The study included retrospective data of 91 children, 55 boys (60.4%) and 36 girls (39.6%) with epilepsy, that were sent by a neuropediatrician for psychological assessment at the Clinical Hospital Center. Except epilepsy diagnosis, another inclusion criterion was that a child had two psychological assessments – one with the RTČ-P in preschool and the other with Wechsler Intelligence Scale for Children – Fourth Edition (WISC-IV) in school age. The average age of the entire sample at the time of the first assessment was 59.63 months (SD = 15.36), and at the time of the second assessment was 108.82 months (SD = 25.46). The average time interval between the two measurements was 49.64 months (SD = 30.05), ranging from 4 months to 12 years. The RTČ-P is specifically designed to assess psychomotor development in children from the first month of life to eight years old. For the purposes of this study, the second part of the test was used, which assesses mental development from ages two to eight. It consists of seven subtests, each with six tasks that include block manipulation, object assembly, pencil manipulation, speech and language development, and verbal knowledge expression. Motor skills are assessed from ages two to four, and tasks to assess verbal knowledge, representing the verbal part of the test, are included from ages five to eight. Based on the child's results, a Mental Development Quotient is determined whereas value above 80 is considered normal, from 60 to 79 is considered borderline, and a quotient of 59 or lower indicates developmental delay. For the second assessment the WISC-IV was used, which is a clinical tool for assessing the intelligence of children aged 6 years to 16 years and 11 months. It measures overall intellectual ability and specific cognitive abilities based on performance on subtests (10 primary and 5 supplementary) and 4 composite scores (Verbal Comprehension Index (VCI), Perceptual Reasoning Index (PRI), Working Memory Index (WMI), and Processing Speed Index (PSI)). These indices contribute to the determination of the Full Scale Intelligence Quotient (FSIQ). The obtained IQ scores are interpreted across categories: a score of 69 or below is extremely low, 70 to 79 is borderline, 80 to 89 is low average, 90 to 109 is average, 110 to 119 is high average, 120 to 129 is superior, and 130 and above is very superior. The study also included demographic data of participants, epilepsy-related data, and data on comorbid conditions. These data were extracted from the participants' medical records. The permission to use the data in this study was approved by the Ethics Committee of the Clinical Hospital Center. Results: In the statistical analisys of the data the following procedures were used: Shapiro-Wilk normality distribution test, descriptive statistics, t-test, correlations and regression analysis. To address the first issue, which is determining the range of results on the RTČ-P in children with epilepsy, the Mental Development Quotient (MDQ) of the entire sample was at an average level (M = 105.45, SD = 16.65). One child (1.1%) showed developmental delay, four children (4.4%) achieved borderline performance, and the majority of participants (N = 86, 94.6%) obtained a normal result (above 80). Regarding specific tasks on the RTČ-P, 17 children (18.7%) failed to solve the ageappropriate graphomotor tasks, 14 (15.4%) did not pass the age-appropriate speech and language tasks, 10 (11%) struggled with block manipulation tasks, while only 5 (5.5%) did not master the age-appropriate object assembly task. More than half of the sample (N = 58, 63.7%) successfully completed all age-appropriate tasks. Therefore, the developmental profile of children with epilepsy in this study is shifted towards higher values, both in terms of the MDQ and in specific developmental areas measured by the RTČ-P. The second problem of this study focused on examining the range of results on the WISCIV in children with epilepsy, both for the FSIQ and for the composite scores (VCI, PRI, WMI, PSI). The overall average FSIQ of the group is 97.53 (SD = 17.76), where 15 participants (16.5%) achieved an IQ score below 80, 12 (13,2%) children achieved low average result, while the highest number of children (N = 38, 41.8%) achieved average scores. High average, superior and very superior result achieved 24 (26,4%) participants. Regarding the composites, there was a shift towards lower values in the domains of Working Memory (M = 94.86, SD = 16.51) and Processing Speed (M = 93.32, SD = 16.52). The highest score was observed in the VCI (M = 102.09, SD = 15.45). The performance on individual subtests is within average range (with scaled scores ranging from 8 to 11), where the lowest score is observed on the Coding, Digit Span and Symbol Search, and the highest on the Similarities. To address the third problem of this study, which aimed to determine the relationship between the quotient of mental development measured in preschool age and the results on the WISC-IV in school age among children with epilepsy, correlations were conducted. The results indicate a statistically significant correlation between the MDQ and FSIQ (r = 0.60, p < 0.001), which is of moderate strength. Similarly, the MDQ is statistically significantly correlated with the results of all composite measures, but most strongly with VCI score (r = 0.62, p < 0.001), while the correlations with other composite measures are lower (r = 0.46, p < 0.001). Although the correlations were found to be statistically significant, from a clinical perspective, it is important to emphasize an extremely wide range of differences in quotients between the two measurements, ranging from -45 to +49 units. Differences of 10 or more units can lead to entirely different conclusions regarding the category of mental/intellectual ability. In the first assessment, the result for 24 children (26.4%) was 10 to 20 units higher than in the second assessment, while for 19 children (20.9%), this difference was greater, ranging from 21 to 49 units. Data was missing for two children (2.2%). In the second assessment, the result for 13 children (14.3%) was 10 to 45 units higher compared to the first assessment. None of the children had the same quotient value in both assessments. To determine the extent to which various participant-related and testing-related data (MDQ, time interval between two measurements, comorbid conditions) and epilepsy-related data (age of epilepsy onset, number of seizures before the first and the second measurement, use of antiseizure therapy during the second measurement) contribute to predicting FSIQ, a regression analysis was conducted. The overall model explains approximately 50.8% of the variance in the criterion variable and is statistically significant. Significant negative predictors of FSIQ variance include the use of antiseizure therapy during the second assessment, the number of seizures before the second assessment and attention and concentration difficulties. Significant positive predictors of FSIQ are the MDQ and the number of seizures before the first assessment which was unexpected because it indicates that higher number of seizures before the first assessment is associated with a higher FSIQ in school-age. Other predictors (time interval between two assessments, age of epilepsy onset, other health problems, emotional difficulties, behaviour difficulties, neurorisk, ADHD) didn't reach statistical significance. Discussion: Studies of children with epilepsy show greater frequency of participants with developmental delays (from 20% to 71%), which is in contrast with our data. One possible reason for this discrepancy could be the specific characteristics of the groups of children with epilepsy that were examined. In the samples from those studies, epilepsy mostly appeared by the age of four, whereas in our sample, epilepsy manifested after the age of four in the majority of children and there are evidence that the period of epilepsy onset up to the age of two is the most critical and is most often associated with greater cognitive impairments and that the risk of cognitive dysfunction decreases in children whose epilepsy appears after the age of five. It is also notable that more than 50% of our sample was tested for the first time after experiencing the first epileptic seizure, i.e., during the diagnosis phase and before the introduction of antiseizure medication. This implies that any developmental/cognitive deficits caused by epilepsy and related factors might not have been significantly expressed yet, and thus could not be detected. It is possible that the RTČ-P is less sensitive to very subtle cognitive changes that may arise as a result of the disease, especially in its early stages. The shift in MDQ towards higher values could also be attributed to other characteristics of the RTČ-P. From a clinical experience perspective, the tasks included in the test are mostly easily solved by children (unless there are very pronounced developmental deficits), and the test itself has not been standardized for about thirty years. All age-appropriate tasks were successfully completed by 63.7% children. Research on the cognitive functions of children with epilepsy generally indicates a wide range of intelligence quotients, with low average and/or below-average group results in certain composites and subtests. These studies highlight reduced results in the areas of working memory and processing speed, with subtest-level data supporting lower efficiency in tasks like Digit Span, Coding, and Symbol Search. These findings are confirmed with our data, too. The best average scaled score for the group, in this study, was achieved in Similarities, which is a subtest commonly identified in literature as one where children with epilepsy perform best. The mentioned range of differences in the results between the MDQ and the FSIQ indicates a high possibility of overestimating the later cognitive status of a child. The risk of underestimating the FSIQ has shown to be lower and, in clinical practice, it might cause less harm. Therefore, it is clear that when drawing conclusions about cognitive and intellectual functioning of children with epilepsy based on the RTČ-P results, it is important to be cautious because the likelihood of misjudgment is high, especially in the terms of overestimating a child's future abilities. The only significant predictor of FSIQ among the comorbid condition variables, are difficulties with attention and concentration. An additional support for this can be found in the participants' performance on the WMI and PSI, the composite scales that require a high level of attention and concentration. Therefore, we can conclude with certainty that these indeed represent one of the cognitive weaknesses in children with epilepsy. It is very interesting that the number of seizures before the first assessment is a positive predictor. Partially, it could be under influence of some epilepsy related variables that we didn't control, like number of antiseizure medication, time interval from the last seizure, seizure type, ect. Also, given that the first assessment was mostly conducted between the ages of 4 and 7, it is worth considering whether the detrimental effects of a higher number of seizures by that age could have been compensated by both physiological processes and environmental stimulation. Neuroplasticity, a special ability of the central nervous system to change the brain's structure and function either temporarily or permanently, is most intense up to the age of 6. We also know that the dynamics and the functioning of a family significantly change when a child has an illness, so it can be assumed that a higher number of epileptic seizures in the early stages of the disease may intensify these changes even more due to the heightened fear of unfavorable cognitive outcomes. If caregivers are much more involved and enable more meaningful/goaldirected developmental stimulation to a child, it can positively affect cognitive maturation. The MDQ of children with epilepsy, measured by the RTČ-P, proved to be a statistically significant positive predictor of FSIQ in school age, consistent with foreign studies using other developmental scales. But, as earlier noted, it is important to keep in mind the limitations of predicting future intellectual functioning based on this developmental test in children with epilepsy. When predicting future cognitive outcomes based on the RTČ-P, it is crucial to adopt a cautious approach and supplement the assessment data with the results from other age- appropriate instruments to ensure a more accurate and reliable prognosis. It is important to remember that it provides a developmental framework rather than a precise developmental assessment on which we can confidently conclude future cognitive outcomes for children with epilepsy. Limitations of this study certainly include the inability to generalize the existing results due to the features of the sample. Specifically, some patients with epilepsy who have more pronounced intellectual impairments could not be included in the study (due to the impossibility of administering the WISC-IV). An additional limitation in drawing conclusions is the lack of a control group of healthy peers. Furthermore, certain demographic variables that have been shown to be relevant in some studies (e.g., SES, parents' education level, etc.) were not included in the research. Moreover, predictive assessment at a single time point certainly cannot encompass all future cognitive and developmental changes, especially considering the variability and heterogeneity of epilepsy itself, along with numerous factors that can positively or negatively impact its course. It would be desirable to have multiple measurements at different time points and to use additional measures of development or intelligence in preschool-age children, as well as behavioral assessments by parents/caregivers. Conclusion: The results indicate that children with epilepsy show average values of MDQ and FSIQ within the average range, with a very wide range of results. Notably, there is a shift towards higher values in the MDQ. It can be concluded that children with epilepsy perform poorest in preschool-age requirements for graphomotor skills, and in school-age requirements for working memory and processing speed. Furthermore, statistically significant correlations were found between the MDQ and all intelligence measures on the WISC-IV, supporting the basis for predicting FSIQ on RTČ-P scores. Significant negative predictors of FSIQ include the use of antiseizure medication during the second assessment, number of seizures before the second assessment, and attention/concentration difficulties. The mental development quotient and the number of seizures before the first assessment were identified as significant positive predictors. To conclude, we can use RTČ-P results as a tool to predict future cognitive outcomes of children with epilepsy, but we need to do that with caution and awareness of it's limitations.