It has been argued that personnel and resources traditionally allocated to special education should be re-directed toward building inclusive schools, partly through working with teachers and systems, and partly through direct work that enables pupil participation (Göransson et al., 2017). Co-teaching is regarded as the best option for this, but is challenged by time constraints and professional inequalities (Mulholland and O’Connor, 2016). In reality, one-to-one support but particularly small group work outside general classrooms are dominant practices, even in countries with well-established inclusive education systems (Paulsrud and Nilholm, 2023). It remains unclear whether such withdrawal, or ‘pull-out’, creates a parallel system of special education within mainstream schooling, or whether it supports inclusion by enabling flexible grouping and a plurality of support practices (Travers and Downes, 2025).

Withdrawal carries the risk of marginalising learners (Buchner and Köpfer, 2025) and teachers note much of what is offered in withdrawal spaces can be provided within class (Vlachou et al., 2015). Despite this, research suggests learners can appreciate separate learning spaces, describing them as quieter, calmer, less pressured and more accessible in terms of instruction (Conn et al., 2024). Some learners express a strong sense of belonging in separate spaces and pride in being part of a distinct group (Holt et al. 2012). These findings reflect wider debates on social inclusion which acknowledge the need to remove barriers while recognising the value of segregated, dignified spaces that individuals play an active role in shaping (Norwich and Webster, 2024).

Belonging has become a central concept for inclusive education in recent years. Children and young people can experience profound exclusion even while physically present in mainstream settings (Dunne et al., 2019), and inclusive education is increasingly understood as requiring meaningful participation and a genuine sense of belonging (Hernández-Saca et al., 2023). Individual learners attend to different aspects of their environment, leading them to experience belonging in deeply personal ways. However, when consulted, they consistently describe belonging as feeling comfortable and safe enough to engage in learning, which they identify as the core activity of their classrooms and schools (Black-Hawkins et al., 2022). Positive peer relationships are crucial for learners, but also positive relationships with teachers: whether they like their teacher, whether their teacher likes them and whether instruction is understandable (Black-Hawkins et al., 2022).

Understanding inclusive pedagogy as a relational process cannot be overstated. Pedagogy does not arise from objects, such as curriculum, learning spaces and tasks, but from interactions between subjects (Gabel, 2002). Teachers must view learners as attentive, capable and agentive, their learning subjectivities emerging out of responsive pedagogical relations. A caring attitude is essential since learning is emotionally situated, but pedagogy is ultimately grounded in knowledge-based relationships (Biesta, 2017). Dialogic practices, such as talking, questioning and thinking together, are central to fostering engagement and learning.

This applies to all learners, including those who bring difference. While such learners may experience greater frustration and anxiety, making care relations especially important, the core pedagogical experience remains a knowledge-based encounter between teacher and learner. Learners are supported primarily through rich language-mediated practices, including explanation, questioning, discussion, visualisation, praise and reassurance (Prediger and Buró, 2024). This raises a key question: given the demands placed on classroom teachers, how challenging is this relational and linguistic work, and to what extent might withdrawal practices respond to a wider set of priorities and needs (Naraian, 2019)? Are withdrawal spaces addressing learner deficits or offering a range of meaningful and responsive spaces for diversity (Travers and Downes, 2025)?

The overarching aim of the review is to examine research that sheds light on the practice of withdrawal, specifically how it is made sense of by those involved. Of particular interest is pedagogy within withdrawal spaces and whether it could be said to support inclusion within the wider setting. To this end, the following research questions are addressed: [1] What are learners’ experiences of withdrawal in primary and secondary schools? [2] How do they experience teaching and learning within withdrawal spaces compared to learning in mainstream classrooms? [3] What meanings of withdrawal emerge in research with teachers, particularly regarding the pedagogical opportunities it affords learners and how it supports inclusion? Relevant academic literature was identified through searches of the British Education Index, ERIC, Scopus, Academic Search Complete and PsycINFO. Search terms were informed by preliminary reading and predefined criteria limited results to English-language studies published between 2009 and 2025. The review focused on Europe as, despite cross-national variation in education systems and terminology, learning support practices show notable similarities (Stentiford et al., 2018). Only studies conducted in mainstream primary or secondary schools were included. Those focusing on special schools, self-contained classrooms, preschools, or higher education were therefore excluded. A total of 3588 articles were identified from database searching, but most of these did not meet the selection criteria or were focused on areas other than education. The eligibility of the remaining studies (n= 48) was assessed through a closer reading of each article, after which a further 27 were excluded. Additional articles were identified from other sources and a total of 23 articles were used for data extraction. A narrative synthesis approach was used to integrate findings from the selected studies. In this method, information from multiple studies is interpreted, organised into themes and examined for areas of convergence and divergence, with disagreements resolved through discussion within the research team (Green et al., 2006). After agreeing on an initial set of codes, we conducted a more detailed reading of each study to extract relevant textual units. These units were then sorted into themes aligned with our research questions. As a result, some studies were coded and themed several times.

Analysis is currently ongoing but emergent findings suggest that across European contexts withdrawal practices are widespread and have become a routine organisational response to perceived learner diversity, particularly as learners get older. A key theme emerging is the tension between inclusive ideals and practical constraints. Teachers generally express commitment to ‘all-in’ inclusion, yet simultaneously view withdrawal as necessary under certain conditions. These conditions include curriculum demands but also the perceived need for quieter or more individualised learning spaces. In several studies, pull-out provision is framed as pedagogically effective, allowing slower pacing, explicit instruction, relational support and reduced performance pressure for learners. However, the findings also highlight significant risks. Withdrawal can construct students as ‘not able’ or ‘not normal’, so supporting ability-based teaching and producing stigmatising spaces that intensify divisions between learners. Spatial arrangements play a key role: where separate rooms become fixed ‘integration’ or ‘support’ spaces, they often solidify hierarchies of ability. In contrast, flexible, multifunctional spaces used by a wide range of learners are seen to operate as enabling rather than marginalising environments. Another recurring theme is the limited collaboration between mainstream and learning support teachers. Across contexts, lack of shared planning time, unclear professional roles and organisational structures favouring withdrawal reduce the potential for inclusive, in-class support. Teachers frequently view specialist expertise as located outside the classroom, which can allow mainstream teachers to displace responsibility for learner diversity. Overall, findings suggest that withdrawal is neither inherently inclusive nor exclusive. Its impact depends on duration, purpose, group composition, coordination with classroom learning and attention to student voice. Reflective use of space emerges as crucial to reducing the segregating effects of pull-out practices while addressing genuine learner needs.

Biesta, G. (2017) The Rediscovery of Teaching. Routledge. Black-Hawkins, K. et al. (2022) Developing inclusive classroom communities Education 3-13 50(5): 577-591. Buchner, T. and Köpfer, A. (2025) Mapping the field: Spatial relations in research on inclusion and exclusion in education IJIE 29(2): 129-136. Conn, C. et al. (2024) Learner experiences of low attainment groups in the context of a rights approach to education Pedagogy, Culture and Society 33(3): 983-999. Travers, J. and Downes, P. (2025) Developing a concentric spatial turn for inclusive and special education European Journal of Special Needs Education. Dunne, L. et al. (2019) Spaces of inclusion: Investigating place, positioning and perspective in educational settings through photo-elicitation IJIE 22(1): 21-37. Gabel, S (2002) Some conceptual problems with critical pedagogy Curriculum Inquiry 32(2): 177-201. Göransson, K. et al. (2017) Ideas about occupational roles and inclusive practices among special needs educators in Sweden Educational Review 69(4): 490-505. Green, et al. (2006) Writing narrative literature reviews for peer-reviewed journals Journal of Chiropractic Medicine 5(3): 101–117. Hannes, K. et al. (2012) Don’t pull me out!? Preliminary findings of a systematic review of qualitative evidence on experiences of pupils with special educational needs in inclusive education Social and Behavioural Sciences 69: 1710-1713. Hernández-Saca, D. et al. (2023) A critical systematic literature review of global inclusive education Education Sciences 13(12): 1212. Holt, L. et al. (2012) Special units for young people on the autistic spectrum in mainstream schools Environment and Planning 44(9): 2191-2206. Mulholland, M. and O’Connor, U. (2016) Collaborative classroom practice for inclusion: Perspectives of classroom teachers and learning support/resource teachers IJIE 20(10): 1070-1083. Naraian, S. (2019) Precarious, debilitated and ordinary: rethinking (in)capacity for inclusion Curriculum Inquiry 49(4): 464-484. Norwich, B. and Webster, R. (2024) Enhancing public dialogue about inclusion in school education Frontiers in Education 9. Paulsrud, D. and Nilholm, C. (2023) Teaching for inclusion - a review of research on the cooperation between regular teachers and special educators in the work with students in need of special support IJIE 27(4): 541-555. Prediger, S. and Buró, R. (2024) Fifty ways to work with students’ diverse abilities? IJIE 28(2): 124-143. Stentiford, L. et al. (2018) A systematic literature review of the organisational arrangements of primary school-based reading interventions for struggling readers Journal of Research in Reading 41(S1): 197-225. Vlachou, A. et al. (2015) Roles, duties and challenges of special/support teachers at secondary education European Journal of Special Needs Education 30(4): 551-564.