The impact of the COVID-19 pandemic on children's and adolescents' development is a topic that has been intensively studied in recent educational research. However, the focus is often on the consequences that school closures and class cancellations had for students (e.g., Betthäuser, Bach-Mortensen, & Engzell, 2023; Patrinos, Vegas, & Carter-Rau, 2022). The consequences that COVID-19 infections can have on affected children and adolescents have been described primarily in clinical studies. Although younger individuals are less likely to have symptomatic infections or severe infections, they may experience symptomatic consequences that are persistent even after recovery (Behnood et al., 2022; Lopez-Leon et al., 2022). The severity of persistent consequences has been linked to the severity of symptomatology during illness (e.g., Radtke, Ulyte, Puhan, & Kriemle, 2021). For adults, such associations have already been empirically demonstrated regarding cognitive impairments (Hampshire et al., 2021). Although many clinical studies examined the consequences of COVID-19 infections in children, there is a lack of studies presenting results that are representative of specific subpopulations and that allow comparisons of groups of children that have already recovered from COVID-19 with groups that have not yet been infected. The present study addresses these research desiderata and examines the question of whether primary school children who have recovered from COVID-19 show disadvantages in terms of their cognitive abilities.

1,761 second- and third-grade students (49.9% girls, 50.1% boys) in the federal state of Burgenland, Austria, who had parental consent to participate were examined in June 2022 (32.3% of the population of students in the school year 2021/22, from 106 (63.1%) of the 168 elementary schools in Burgenland). In addition, students’ parents or legal guardians were surveyed (n=1,438). The key independent variable was whether the children had been infected with COVID-19 at the time of the survey. According to parents and students, this was the case for n=1,253 students, whereas n=508 students had not been infected by that time. In addition, we assessed whether the infection was symptomatic or asymptomatic, which symptoms occurred, and whether a physician was consulted due to the COVID-19 infection. In addition, characteristics of students' individual and family background were surveyed (including gender, language spoken at home, native language, parents' country of birth, parental education, etc.). We used standardized instruments of the federal state of Burgenland to weight the gathered data based on state statistics. As dependent variable, cognitive ability was assessed using the Cognitive Abilities Test (KFT 1-3; Heller & Geisler, 1983) (test duration: 60 minutes), which consists of four subtests: language comprehension, relation recognition, inductive reasoning and numerical thinking. For analysis, four groups were distinguished: children who had not been infected at the time of the survey (control group, n=502) and three recovery groups (RG): asymptomatically infected children (RG1, n=251), symptomatically infected children (RG2, n=850), and symptomatically infected children who had seen a medical doctor because of the illness (RG3, n=131). The doctor's visit is considered as an indicator of a situation that gave the parents reasons for concern. According to parents, 78 percent of recovered children had been infected with COVID-19 within the five months prior to data collection. The data of the three recovery groups were compared pairwise with those of the control group. Since small, but significant differences were found between the groups regarding immigrant background, native language and language spoken at home, an analysis of covariance was conducted controlling for these variables. Missing values were treated as Missing at Random and were multiply imputed (MICE, Buuren & Groothuis-Oudshoorn, 2011; CART Breiman, et al, 1984). Data were weighted using iterative proportional fitting (IPF; Deming & Stephan, 1940; Lomax & Norman, 2019) based on representative statistics from the federal state of Burgenland. All statistical tests were conducted with an error probability of p<.05.

Regarding cognitive abilities, RG1 and RG3 showed a significantly lower test performance in numerical reasoning than the control group (RG1: F(1,748)=7.42**, p=.007, partial Eta²=.010; RG3: F(1,627)=9.18**, p=.003, partial Eta²=.014). Moreover, RG3 also performed significantly lower in language comprehension than the control group (F(1,627)=11.26***, p<.001, partial Eta²=.018). For relation recognition and inductive reasoning, RG3 performed, in tendency, lower than the control group (F(1,627)=3.57, p=.059, partial Eta²=.006; F(1,637)=3.19, p=.075, partial Eta²=.005). Our findings suggest negative cognitive effects of COVID-19 infections for two of the recovery groups distinguished in the present study. For the recovery group of symptomatically infected children who underwent medical treatment, the findings point more strongly into this direction. The identified effects are of small size. However, given the low prevalence of longer-lasting symptoms after the infection among children (Lopez-Leon et al., 2022), these effects may imply severe consequences for the cognitive functioning of the respective children. Further analyses using propensity score matching are planned to validate our findings obtained by covariance analysis. Beyond this, it has to be considered that the effects reported here emerged at a time when, for most children, the infection happened only a few weeks or months before the survey. The extent to which these effects persist is another important question. Therefore, our sample was re-assessed in June 2023 using the same test instrument to assess students' cognitive abilities. The results of this follow-up study will be available by spring 2024 and will be included in our paper. The findings will be discussed with reference to the medical research literature as a consequence of the impaired central functions (memory, attention) and with regard to consequences for targeted educational support of children after their COVID-19 infections.

Behnood, S. A., Shafran, R., Bennett, S. D., Zhang, A. X. D., O’Mahoney, L. L., Stephenson, T. J., . . . Swann, O. V. (2022). Persistent symptoms following SARS-CoV-2 infection amongst children and young people: A meta-analysis of controlled and uncontrolled studies. Journal of Infection, 84(2), 158–170. https://doi.org/10.1016/j.jinf.2021.11.011 Betthäuser, B. A., Bach-Mortensen, A. M., & Engzell, P. (2023). A systematic review and meta-analysis of the evidence on learning during the COVID-19 pandemic. Nature Human Behaviour, 7(3), 375–385. https://doi.org/10.1038/s41562-022-01506-4 Breiman, L., Friedman, J. H., Olshen, R. A., & Stone, C. J. (1984). CART: Classification and Regression Trees. Belmont, CA: Wadsworth. Buuren, S. van, & Groothuis-Oudshoorn, K. (2011). mice: Multivariate imputation by chained equations in R. Journal of Statistical Software, 45(3), 1–67. https://doi.org/10.18637/jss.v045.i03 Deming, W. E., & Stephan, F. F. (1940). On a least squares adjustment of a sampled frequency table when the expected marginal totals are known. The Annals of Mathematical Statistics, 11(4), 427–444. https://doi.org/10.1214/aoms/1177731829 Hampshire, A., Trender, W., Chamberlain, S. R., Jolly, A. E., Grant, J. E., Patrick, F., . . . Mehta, M. A. (2021). Cognitive deficits in people who have recovered from COVID-19. EClinicalMedicine, 39, 101044. https://doi.org/10.1016/j.eclinm.2021.101044 Heller, K., & Geisler, H. J. (1983). Kognitiver Fähigkeitstest (Grundschulform). KFT 1–3. Weinheim: Beltz. Lomax, N., & Norman, P. (2016). Estimating population attribute values in a table: “Get me started in” Iterative Proportional Fitting. The Professional Geographer, 68(3), 451–461. https://doi.org/10.1080/00330124.2015.1099449 Lopez-Leon, S., Wegman-Ostrosky, T., Ayuzo del Valle, N. C., Perelman, C., Sepulveda, R., Rebolledo, P. A., . . . Villapol, S. (2022). Long-COVID in children and adolescents: A systematic review and meta-analyses. Scientific Reports, 12(1), 9950. https://doi.org/10.1038/s41598-022-13495-5 Patrinos, H. A., Vegas, E., & Carter-Rau, R. (2022). An analysis of COVID-19 student learning loss. The World Bank. https://doi.org/10.1596/1813-9450-10033 Radtke, T., Ulyte, A., Puhan, M. A., & Kriemler, S. (2021). Long-term symptoms after SARS-CoV-2 infection in children and adolescents. JAMA, 326(9), 869–871. https://doi.org/10.1001/jama.2021.11880