In the preliminary phase, a scoping search was carried out, by the guidelines of Robinson et al. (2014), to identify SLR, bibliometrics, and/or scientometrics of the TPACK, from 2020 onwards. The following databases and/or repositories were included: Dialnet, Google Scholar, Scientific Journals of Latin America and the Caribbean, Spain and Portugal (REDALYC), Semantic Scholar, and Scientific Electronic Library Online (SciELO). The keywords included in the search were "technological AND pedagogical AND content AND knowledge AND TPACK" and adaptation according to the particularities of each database. Thirty-six researches were obtained, and the scientific production is preferentially concentrated in the year 2023 with 55,55% of the total. The most popular databases and/or repositories are SCOPUS, Google Scholar, Web of Science (WoS), and Education Resources Information Center (ERIC). To the subjects addressed, there are studies related to the subjects of mathematics, science, English, and geography and other subjects specific to local realities such as China, Indonesia, and Ibero-America. Finally, the topics of e-learning, Science, Technology, Engineering, and Mathematics (STEM), teacher training, and primary education stand out, see Table 5, Appendix.

In consideration of the previous background, an SLR was developed including the following databases and/or repositories: DIALNET, ERIC, JSTOR, PsycINFO, REDALYC, SCOPUS, and WoS. Keywords were checked in the ERIC and UNESCO thesauri. The search equation included the term TPACK, as it provides a broader view, avoiding focusing on one or two components (Mishra and Koehler, 2006; De Rossi and Trevisan, 2018), see Table 2. The search period covered was from the inception of the model 2006 to October 2023.

Table 2
Specific keyword protocol in each repository.

Source: Own elaboration.

Inclusion criteria were: articles, open access, full text, social sciences, and research developed in special education contexts. Exclusion criteria included: abstracts, editorials, press releases, conference papers, master's and doctoral dissertations and theses, areas other than social sciences, and research that did not consider special school contexts.

Figure 2
Summary outline of selected articles
Source: Own elaboration.

Figure 2 shows 286 documents documents were obtained in the identification stage, the largest number corresponding to DIALNET (44,02%), followed by JsTOR (35,07%). The articles were reviewed according to titles, keywords, and abstracts, by the inclusion criteria. In addition, in some cases, access to the full text was required. Finally, 16 articles were selected, seven SCOPUS (43,75%), six PsyINFO (37,5%), and one each in ERIC, JsTOR, and WoS (6,25%), see Figure two. Systematic reading of the articles was carried out to obtain information related to the research questions, see Table 3.

Table 3
Research articles included in the SLR

Source: Own elaboration.

Publications are distributed between the period 2013 and 2023, with 2021 being the year of highest production, accounting for 25% of the total, see Figure 3.

Figure 3
TPACK articles by publication year
Source: Own elaboration.

In terms of geographical distribution, the United States is the country with the highest scientific production, with 50% of the articles reviewed. It should be noted that the European context presents a participation of 11 different countries.

Regarding the type of research, 37,5% are mixed, followed by 25% quantitative and 18,75% design-based research and other types of studies, see Figure 4.

Figure 4
Types of research found in this SLR
Source: Own elaboration.

The research is carried out mostly in university education with 56,25%, followed by primary and secondary education with 12,5% ​​respectively. The most representative samples correspond to Karaaslan et al. (2023) and Barbieri et al. (2021) with 446 and 432 students from Turkey, Italy, Greece and Portugal, respectively. On the other hand, the smallest sample of only three teachers (Ciampa, 2017). Regarding the instruments, there is a predilection for the use of questionnaires that represent 43,75% of the total, followed by field notes with 18,75%. Finally, among the research topics or themes, mathematics stands out with 37,5% of the total, followed by the Braille language and science.

The second part of the analysis considers some of the approaches made by Paidicán and Arredondo (2022a) to provide answers to questions two, three, and four.

Table 4
TPACK research in special education by level of education

Source: Own elaboration.

Table 4 shows that more than half of the studies of the TPACK model in special education are carried out at the university level, followed by primary and secondary with 12,5% respectively. In the primary studies, they present qualitative methodologies and case studies, with samples ranging from three to 80 cases, including students and teachers. Both studies consider three or more instruments including field notes, reflections, interviews, document analysis, observations, and student work. The subjects of educational technology and language through Braille were considered. Finally, the studies were carried out in the United States and South Africa. Referring to the findings, teachers require strong PK, TK, and CK knowledge, which needs to be continuously addressed in professional development processes. In addition, teachers require adequate time to explore and critically evaluate digital resources to strengthen personalization, taking into account the particularities of learners, for example in instances of visual impairment and the use of Braille

About the recommendations, there is a need to opt for participatory action research approach models (McTaggart, 1991) and active learning techniques (Desimone, 2009; Jaipal-Jamani & Figg, 2015), to ensure collaboration and social participation (Ciampa, 2017). To secondary school studies, there are mixed research and consensus statement processes, with samples ranging from five to 125 teachers. The instruments used were the TPACK-21 questionnaire, self-efficacy scale (SE-PETE-D), interviews and logbook. The areas included in the research were physical education and mathematics using Braille. The studies were conducted in Europe, with the research by Kwok et al. (2021) including up to five countries.

The results indicate that teachers require greater TK and CK, both for special education and in mathematics and physical education, favoring the development of classes with appropriate adaptation according to the characteristics and context of the students. Finally, teachers need to be more aware of the importance of student support, such as, for example, the use of Braille. About the recommendations, more space for reflection among teachers is required, including more representative samples, to deepen the aspects related to the European Standards in Adapted Physical Activity (EUSAPA). Also, aspects related to the use and selection of assistive devices for students with SEN should be deepened.

Regarding the studies focused on the university setting, they mostly present design-based, qualitative, and mixed methodologies, which show an average of 129 cases, with the study by Karaaslan et al. (2023) presenting the most significant sample with 446 students, more than 60% of the studies were developed in the United States, including undergraduate and graduate students. The most commonly used instruments were the TPACK questionnaire and rubrics. The study by Anderson et al. (2017) presented the largest variety of instruments including: focus group discussion, focus group, interviews, lesson plans, field notes, and observations.

The results indicate that teacher training in areas related to TPACK facilitates the incorporation of IPAD and DYNABOOK, improving learning experiences and the systematic integration of ICT, including important areas such as logical-mathematical reasoning (Anderson et al., 2017; Courey et al., 2015; Kuo, 2015). Furthermore, the levels of self-efficacy measured utilizing the TPACK questionnaire indicate that women perform better than men and that the use of this type of instrument is valid for postgraduate students (Kaplon-Schilis and Lyublinskaya, 2019; Karaaslan et al., 2023). Also, it was observed that the use of the TPACK model allowed for specific didactic insights in areas related to E-learning, Braille, artificial intelligence, augmented reality, and 3D printing (Kotzebue et al., 2021; Lee et al., 2014). Finally, the use of new tools such as the rubric allows for a broader view of the TPACK model, including both theoretical and practical aspects, as exemplified by the analysis of lesson plans (Lyublinskaya and Tournaki, 2014; Lyublinskaya and Kaplon-Schilis, 2022).

In relation to the recommendations, the consolidation of new instruments related to TPACK is required, in different contexts and with larger samples (Courey et al., 2015; Lyublinskaya and Kaplon-Schilis, 2022). There is also a need to deepen lines of research related to classroom design, classroom culture and organisation, and the use of ICT and their ways of thinking and acting at the time of implementation (Anderson et al., 2017; Kotzebue et al., 2021; Lee et al., 2014). Finally, it requires the development of training processes in special education settings, including teacher and students, as well as the use of feedback as a mechanism for teachers' professional reflection (Kuo 2015; Lyublinskaya and Tournaki, 2014).

Studies that include more than one educational level, are characterized by their mixed methodology, with samples ranging from 60 to 432 individuals, the research by Barbieri et al. (2021), which includes the participation of students from Greece, Italy, and Portugal, stands out. The preferred instruments are the TPACK Questionnaire (Zelkowski et al., 2013) and interviews. The results indicate that high teacher confidence positively affects ICT integration, even though time and resources are the main obstacles (Hill, 2020). Moreover, the incorporation of Serius Game promotes significant changes, especially in mathematics, and, in addition, activates mental resources (Barbieri et al., 2021). It is recommended that future research should delve deeper into classroom observations, as they allow for a more objective view of ICT integration through TPACK.

Finally, the study developed in special schools sought to compare the realities of Chinese and Taiwanese teachers with a sample of 415 schools, the instrument used was the TPACK questionnaire (Chai et al., 2011). The results indicate that Chinese teachers perform better than their Taiwanese peers, the likely explanation being that China has educational policies related to special education. It is recommended that a wider variety of instruments be applied during data collection.