Attachment Theory for Teachers: Becoming a Secure Base

As we open our classroom doors each morning, we're managing more than 30 distinct nervous systems. Some arrive calm and ready; others are already in crisis mode before the bell has rung. The difference often comes down to one invisible factor: attachment.

Attachment theory is the framework, set out by John Bowlby (1969) and tested empirically by Mary Ainsworth (1978), which explains how a child's earliest relationships with caregivers build an internal working model: a durable expectation of whether adults can be relied on when the child is distressed. That model travels into your classroom. It is the difference between a learner who puts a hand up when a task is too hard and one who tears the sheet up instead.

Bowlby (1969) showed that attachment shapes how learners relate to others and how they learn. This is not work for psychologists alone. Teachers need to know it too, and most of us have had little training in it.

Without that training, we often read the behaviour wrongly (Ainsworth, 1978; Main & Solomon, 1990). Prior and Glaser (2006) explained that these behaviours are survival strategies.

‍

‍

What Is Attachment Theory?

At its core, attachment theory proposes a simple but profound truth: human beings are wired to seek proximity to a primary caregiver, because doing so increases our chances of survival. This isn't sentimental. From an evolutionary perspective, a child separated from their caregiver is a child in mortal danger.

Our brains have therefore evolved with an invisible tether, keeping us close to the adults responsible for our care. Attachment sits alongside the other fundamental theories of learning that shape classroom practice, though it describes relationships rather than instruction.

Bowlby (1969) built the theory from wartime observation. Working with children separated from their families during and after the Second World War, he saw that institutional care could meet every physical need and still leave children badly harmed. What was missing was not food or shelter. It was a consistent person, and no amount of material provision substituted for one.

That work now sits inside UK policy rather than outside it. Attachment thinking runs through children's social care, virtual school practice and Department for Education guidance on supporting children in care, which is why the language turns up in your safeguarding briefing as well as in the research literature.

Attachment-aware practice has also moved into teacher education and CPD (Prior & Glaser, 2006), and a growing number of schools now describe themselves as attachment-aware settings (Bomber, 2007). The label is easy to adopt. The whole-staff consistency behind it is not.

But what does attachment actually do in the brain? Attachment is not primarily about love or affection, although those feelings often accompany it. Instead, attachment is about the regulation of the nervous system. When a caregiver responds consistently and predictably to a child's needs, the child's brain learns that the world is safe and that they can trust adults to help them.

This creates what neuroscientists call "co-regulation", where the child's stressed nervous system is calmed by the caregiver's calm presence. Over time, the child internalises this calming presence and develops the ability to regulate their own nervous system. This process is called auto-regulation, and it is essential for learning.

This survival response interferes with their learning and behaviour (Bowlby, 1969). Securely attached learners are ready to learn and ask for help in class. They cope with mistakes and feelings without activating a fear response. Learners lacking secure attachment struggle as their brains constantly scan for danger (Main & Solomon, 1990).

‍

Four Stages of Attachment Development

Bowlby identified a progression through which typical attachment develops. Understanding these stages helps us recognise where a child might be stuck and what experiences they may need to recover lost ground.

Stage 1: Pre-attachment (0-6 weeks). Newborns produce behaviours such as crying, sucking, and grasping that encourage adults to respond to them. However, they do not yet show a preference for a particular caregiver. Any adult can soothe them.

Stage 2: Attachment-in-the-making (6 weeks-6 months). The infant begins to prefer familiar caregivers but will still accept comfort from others. They produce more directed smiling and show reduced crying when handled by people they know.

Stage 3: Clear-cut attachment (6 months-3 years). The child now shows obvious preference for their primary attachment figure and becomes distressed when separated from them.

They use the caregiver as a "secure base" from which to explore the world. If the caregiver is present, the child feels confident enough to play, investigate, and take (safe) risks. When the caregiver leaves, the child becomes anxious.

Stage 4: Goal-corrected partnership (3 years onwards). The child starts to grasp that the caregiver has feelings, plans and reasons of their own. Separation can now be negotiated rather than simply protested against (Bowlby, 1969).

By this point the child has built what Bowlby (1969) called an internal working model: a mental template of what adults do when you need them. Main et al. (1985) showed that these models persist, shaping behaviour and later relationships well beyond early childhood.

Learners often reach stage 4, but those with neglect may stay in stage 2 or 3. This presents as clinginess or separation distress (Bowlby, 1969; Ainsworth, 1978). Knowing this lets you respond patiently, not with frustration.

‍

Strange Situation Experiment and Attachment Styles

Ainsworth et al. (1978) designed the Strange Situation. In this controlled test, researchers watched how babies reacted when a caregiver left the room and then came back. What a baby does at the reunion is the key signal. It links that behaviour to how available the caregiver has been, and to how safe the baby feels.

In the Strange Situation, researchers watched three things: how babies explored, how they coped when the caregiver left for a short time, and what they did when the caregiver came back. Ainsworth et al. (1978) named three patterns: secure, avoidant and ambivalent/resistant.

Main and Solomon (1990) later added a fourth: disorganised/disoriented. Here the baby has no clear plan to fall back on. Their responses show conflict or fear instead.

The reunion episodes matter because they show whether the child can use the caregiver as a secure base. Secure infants may become upset by separation but seek contact and settle after reunion. Insecure patterns can involve avoidance, resistance to comfort or confused behaviour, so teachers should read classroom responses as possible signals of relational stress rather than defiance alone.

Ainsworth et al. (1978) watched babies with their carers. They did not study learners in classrooms. So teachers should not label a child based on classroom behaviour alone.

The safer read across is a modest one. Some learners need routine, repair after a fallout, and an adult who is there for them. Only then can they risk being wrong in a lesson.

‍

Four Main Attachment Styles Explained

Ainsworth et al. (1978) described three patterns in the Strange Situation: secure, avoidant and ambivalent/resistant. Main and Solomon (1990) later named disorganised/disoriented behaviour. These are research groups, not quick labels for your class. Even so, they help you think about how safe a learner feels.

Secure attachment happens with reliable caregivers (Bowlby, 1969). Secure learners explore well and trust adults. Learners learn to soothe themselves.

They easily ask for support and make friends (Ainsworth et al., 1978). Learners also bounce back from problems (Sroufe, 2005).

Care that changes from day to day can lead to anxious attachment. Learners never know how the adult will react (Bowlby, 1969). So they watch adult moods closely and stay near.

These learners may cling to teachers (Ainsworth et al., 1978). Transitions are hard, and they react to perceived rejection. Fear of abandonment impacts trust after separation (Main & Solomon, 1990).

Bowlby (1969) linked caregiver unavailability with avoidant attachment. Learners learn to suppress needs, seeing comfort as pointless. In class, they might refuse help and avoid eye contact.

Cassidy & Berlin (1994) noted learners resist comfort and disconnect from peers. This behaviour, Main & Solomon (1990) explain, is a defence, not confidence.

Disorganised attachment arises from frightening care (Main & Solomon, 1990). The caregiver should provide safety, but becomes a source of fear. This creates conflict for the learner, who needs a dangerous caregiver.

Learners may freeze, fear the caregiver, or act aggressively (Lyons-Ruth, 1996). They often show challenging behaviour at school (Boris & Zeanah, 2005). Disorganised attachment links to trauma and abuse (van IJzendoorn et al., 1999).

‍

How Attachment Affects Learning

Attachment theory and classroom research agree on one thing: learners need to feel safe with adults. Roorda et al. (2011) found that warm teacher-learner bonds link to better engagement and results. Bergin and Bergin (2009) show how attachment ideas can shape the bonds you build. None of this proves that attachment alone decides how well a child learns.

A safer classroom implication is to reduce threat and make adult help predictable. When learners know what will happen next, who will help them and how mistakes will be handled, they are more likely to use attention on the task rather than on monitoring adult reactions.

Some learners have lived with unsafe relationships or with trauma. They can read a neutral look or tone from an adult as a threat. Answer that with routines they can predict, calm correction and repair afterwards. If your concerns carry on, bring in safeguarding, SEND or mental-health staff.

Attachment is one factor among many. Weigh it up alongside SEND, language, access to the curriculum, sleep, attendance, mental health, stress at home and school relationships. Never treat attachment as the only reason for slow progress or hard behaviour.

When a learner feels safe with you, it is easier for them to take risks in their work and to work with others. But avoidance, clinginess or anger is not proof that a child sits in one attachment group. Note the patterns you see and adapt your support. Ask a specialist when behaviour points to trauma, safeguarding risk or unmet SEND.

Past experience and how safe a learner feels now can shape how they get on with peers. Give each child a clear role in group work, keep routines steady, and step in as an adult to help where needed.

Keep your thinking broad, though. Communication needs, anxiety, autism, ADHD, language delay and social learning may all play a part.

‍

Identifying Attachment Issues in Learners

Bowlby (1969) argued that attachment needs surface in ordinary daily behaviour, and you don't need psychology training to spot a learner who is struggling. What helps is knowing which behaviours to watch, and the patterns Ainsworth et al. (1978) described give you a starting point.

Anxious attachment often looks like a constant bid for attention. Watch what happens when you are busy with someone else: does the learner escalate, or go visibly flat? Transitions tend to be hard for them too (Bowlby, 1969; Ainsworth, 1978).

Avoidant attachment might show as a learner avoiding eye contact and your presence. They may refuse help or get defensive when you support them. Do they seem disconnected, preferring solitude in class? Minimal emotional response to praise or criticism is also a signal.

Main and Solomon (1990) said watch for fear or aggression in a learner. Look for dissociation, blank stares, or dazed movements. Lyons-Ruth (2003) showed some learners struggle to link actions and outcomes.

Crittenden (1994) noted anger far bigger than the trigger. Cicchetti (2016) tied these signs to disorganised attachment after trauma.

Some children also show age-inappropriate clinginess or extreme reluctance to separate from a parent at drop-off, even in reception or year 1. Others show unexplained aggression towards adults who are kind to them, or freeze when comforted. These are not character flaws or "difficult" behaviour; they are signals that the child's attachment system has not had its needs met.

Learners with attachment issues might have autism or ADHD. Speech delay or selective mutism may also occur (Bowlby, 1969). Insecure attachment does not change diagnoses, yet it complicates them (Ainsworth, 1978; Main & Solomon, 1990).

‍

Teachers as Secondary Attachment Figures

Children do not form one attachment and stop there. Bowlby (1969) and Ainsworth (1978) both described a hierarchy of attachment figures, so a secure relationship with a teacher sits alongside the relationships at home rather than competing with them. Howe (2006) is clear on this, and the worry teachers sometimes carry about overstepping is misplaced.

‍

Bowlby (1969) showed that secure bonds help learners. A learner can build these bonds with more than one adult (Ainsworth, 1978).

Teachers and grandparents can both play this role (Main & Solomon, 1990). Each bond adds something of its own, and none of them weakens the rest.

You are not replacing parents; you offer learners a secure base at school. You provide six hours of predictable, consistent support. This may be the learner's most reliable safety source if home is chaotic (Bowlby, 1988; Ainsworth, 1969).

Becoming a secondary attachment figure does not require grand gestures. It requires the persistent micro-moments of connection. It is learning a child's name and using it consistently. It is greeting them warmly every morning, even (especially) on the day after they have been rude or aggressive to you.

It is noticing small details about their lives and asking genuine questions. It is keeping your promises: if you say you will read to them after lunch, you do it. It is repairing ruptures when you lose your patience or respond harshly.

Bowlby (1969) showed secure attachment builds trust. Anxious learners learn you are reliable (Main & Solomon, 1990). This lessens their worry.

Ainsworth et al. (1978) found avoidant learners see your help as new. They learn support comes without pressure to perform.

None of this works if staff are running on empty. A teacher under sustained pressure has less capacity to be the calm, predictable adult the theory asks for (Bowlby, 1969), and adult stress shows up quickly in learner behaviour (Perry, 2009). Schools that adopt attachment-informed practice across the whole staff see better outcomes than those that leave it to individual goodwill (Bomber, 2007).

‍

Creating Attachment-Aware School Environments

Bombèr and Nursten (2016) highlight the importance of attachment styles. Geddes (2006) offers practical ways to build secure attachments. Perry (2009) shows this improves learner wellbeing and progress. Schools support learners who have adverse childhood experiences.

An attachment-aware school is not a specialist setting. It is a mainstream school that has made a deliberate, whole-system commitment to understanding and supporting children's attachment needs. What does this look like in practice?

Attachment-aware schools see behaviour as communication; Bomber (2023) explores the learner's nervous system. They look at needs, not defiance, says Bomber (2023). This shifts practice to relationship-based methods.

Attachment-aware schools run on routine. Teachers greet every learner by name each day (Bomber et al., 2021). They spell out what happens next, and when.

Signposted transitions are calm. Learners know expectations are visible (Golding & Hughes, 2012).

Third, attachment-aware schools recognise the central importance of a named adult. Often, this adult is not the main teacher but rather a teaching assistant, mentor, or key person. This adult's role is to build a consistent relationship with the child, to check in with them regularly, and to notice subtle shifts in wellbeing or behaviour that might otherwise be missed.

Fourth, attachment-aware schools build time for connection into the school day. This might look like a "settling circle" at the start, a quiet check-in time during lunch, or structured one-to-one time with key adults. These moments of focussed, attuned attention are not luxuries; they are essential for children with attachment difficulties.

Fifth, attachment-aware schools use "emotion coaching", a technique where adults help children name and process their feelings in real time. Rather than saying "don't cry," an emotion coach says "I can see you're feeling sad about leaving the activity. That's a big feeling. Let's take some deep breaths together." Over time, children learn that feelings are normal, manageable, and not something to be ashamed of.

Attachment-aware schools look after staff first. When staff feel stressed, it is harder for them to be a steady adult for a learner.

So these schools offer supervision, training and a sense of safety. That support helps staff stay calm and present for the children in front of them (Bowlby, 1969; Ainsworth, 1978; Main & Solomon, 1990).

‍

Practical Strategies for Supporting Insecurely Attached Children

If you teach a child with attachment difficulties, you are not powerless. Whilst you cannot "fix" a child's early history, you can provide experiences that gradually reshape their internal working models and nervous system responses.

Create a consistent welcome ritual. Greet the child by name every single morning. The consistency matters more than the words.

Some teachers use a handshake, a specific greeting, or a five-second one-to-one check-in. This tells the child: I notice you, I'm glad you're here, and you're safe with me.

Narrate transitions explicitly. Anxiously attached children particularly struggle with transitions because they experience every change as a threat to their safety. Instead of saying "time for PE," say: "We've finished phonics. I'm putting the phonics books away.

In two minutes, we're going to walk to the PE lesson. Mrs Chen will be there waiting for us. After PE, we'll come back here and I'll read to you at 2pm." This detailed narration reduces anxiety by making the invisible visible.

Use "connection before correction." Before addressing a behaviour, make a brief emotional connection. Get down to the child's eye level if safely possible. Use a calm tone.

Acknowledge what you see: "I can see you're feeling really frustrated right now. I want to help." This simple act of recognition and attunement often stops escalation in its tracks.

Embrace the power of repair. You will sometimes lose your temper, snap at a child, or respond harshly.

Rather than pretending it didn't happen, repair it. Find the child later and say: "I'm sorry I raised my voice earlier. I was frustrated, but that wasn't okay.

I'm glad you're in my class, and I care about you." This is not weakness; it is profoundly teaching. For a child with insecure attachment, experiencing a rupture that is successfully repaired is deeply significant. It teaches them that relationships can survive conflict, that adults take responsibility for their behaviour, and that they are valued even when imperfect.

‍

‍

Low-demand time builds relationships. Shared activities like colouring can help. Follow the learner's lead to show interest. For learners with avoidant attachment, this connection builds trust (Bowlby, 1969; Main & Solomon, 1990).

Meet dysregulation with calm. When a child is dysregulated, screaming, hitting, or having a meltdown, their nervous system has shifted into survival mode. Reasoning, punishment, or raised voices will not work; they will only confirm the child's fear that adults are threatening. Instead, stay calm, keep your voice quiet and slow, and focus on safety.

The goal is not to punish the behaviour in that moment; it is to help the child's nervous system return to a state where learning is possible. Repair and reflection can happen later, when the child is regulated.

Build in "predictable surprises." Predictability carries most of the weight here, but a small amount of planned novelty helps. If a child knows they might get a special treat or do something fun on Fridays, this creates positive anticipation rather than anxiety about the unknown. The surprise is predictable because the child has learned: good things happen sometimes, and that's safe.

‍

Attachment Theory and Behaviour Policies

Zero-tolerance behaviour policies were not designed with these children in mind. Attachment theory (Bowlby, 1969) sits awkwardly against an escalating sanction ladder, because exclusion and public shame land hardest on the learners least able to absorb them (Prior & Glaser, 2006). Securely attached learners tend to cope with a strict system (Hughes, 2006). Insecurely attached learners often do not.

When a child with anxious attachment is excluded from school for breaking a rule, the message their nervous system receives is: "You are unsafe, you have been rejected, and the adult has withdrawn care." This confirms their deepest fear and makes their attachment system even more dysregulated. They return to school more anxious, more clingy, or more aggressive, not less.

Similarly, when a disorganised child is harshly punished for behaviour driven by trauma responses, the punishment does not teach cause and effect. Instead, it re-traumatises. The child does not think, "I won't do that again." The child thinks, "Adults are scary, and I need to protect myself."

An attachment policy builds relationships and helps mend them. Restorative justice helps learners face the harm they caused better than punishment does (Bowlby, 1969; Ainsworth, 1978). Insecure learners can learn how their actions land on other people (Bomber, 2007; Perry, 2009).

This is not permissiveness. Boundaries remain clear, and consequences still exist. But the consequence is designed to teach, reconnect, and help the child develop better strategies.

A child who hits a peer is not sent out of the room; instead, a trusted adult sits with them, helps them understand what triggered the response, and coaches them through what they could do differently next time. Over time, this builds capacity and resilience far more effectively than exclusion ever could.

‍

Attachment Theory and Childhood Trauma

Adverse Childhood Experiences (ACEs) are now well documented in UK cohorts. Abuse, neglect and household dysfunction affect development in measurable ways (Anda et al., 2006), and higher ACE scores are associated with poorer health and weaker educational outcomes across the life course (Felitti et al., 1998).

Attachment is the lens through which we must understand ACEs. Many ACEs directly disrupt attachment formation. A child who has experienced abuse or neglect has not learned that adults are safe.

A child who has been moved through multiple care placements has experienced repeated severing of attachment bonds. A child whose parent is addicted or mentally ill may have experienced profound inconsistency and unpredictability.

Looked-after children and learners facing hardship need support. UK law protects children in care. Virtual school heads oversee their education.

Designated teachers and pupil premium funding aim to improve outcomes. Attachment theory (Bowlby, 1969) shows these learners need trust rebuilt with adult support.

Trauma training helps teachers support learners in care, largely because it changes what you see. Hypervigilance and sudden aggression read as trauma responses rather than defiance. Uprooting events such as a house move or a change of school can destabilise these learners fast (Perry, 2006), because their sense of safety was fragile to begin with (van der Kolk, 2014).

Talk with carers and the Virtual School. Looked-after learners need consistent support from everyone. Use the same behaviour strategies in school and at home. Regular contact stops learners facing different rules (Bomber, 2011).

‍

‍

Criticisms and Limitations of Attachment Theory in Education

Attachment theory is influential and well-supported by research, but it is not without limitations or valid criticisms. Teachers should be aware of these. Use it as a starting point for professional discussion: identify the learner's current need, record evidence from more than one lesson, and agree the next classroom adjustment with the SENCO or family.

Overdiagnosis is a risk. Some learners are shy, which is normal. Others process loss (Bowlby, 1969). Teachers might pathologise usual behaviour (Prior & Glaser, 2006) after learning about attachment (Ainsworth, 1978).

Ainsworth (1978) showed that culture shapes how we read attachment. The way a family cares for a child changes how that bond looks in class. Some cultures value close family ties over standing alone (Bowlby, 1969). So keep family life in mind when you judge a learner's bond with adults.

Workload is the second criticism, and it is a fair one. Teachers asked to provide therapeutic support alongside a full timetable will be overloaded, and attachment awareness can quietly become a route to importing social work into the classroom without importing the training, supervision or time that goes with it.

Teachers can be aware, but are not therapists (Prior & Glaser, 2006). Schools need specialists for complex trauma, not just teacher training (Perry, 2009). Attachment theory should not justify extra work without resources (Bomber, 2007).

Attachment theory helps explain learning issues, but isn't the full story. Learners with secure attachments can still have dyslexia or anxiety. Attachment matters, but assess all learner needs thoroughly (Bowlby, 1969; Ainsworth, 1978).

‍

‍

Key Takeaways for Teachers

Attachment is about nervous system safety, not just emotion or affection. A securely attached child is one whose brain has learned that the world is predictable and that adults can be trusted to help. This allows them to focus their cognitive energy on learning rather than threat-detection.

Children develop survival strategies based on early experiences (Bowlby, 1969). These strategies aren't flaws but adaptations to feel safe. A learner's behaviour shows their learned way of coping (Ainsworth, 1978). Understand this, and respond to learners with compassion, not judgement (Prior & Glaser, 2006).

Teachers can offer learners a secure base, aiding their learning (Bowlby, 1988). You don't need costly actions or therapy skills (Rutter, 1985). Just be consistent, pay attention and make small, regular connections (Pianta, 1999). Your reliable presence is important (Hughes, 2006).

Insecure attachment may be one reason a learner finds it hard to trust adults, work alone or take correction. But it is never the whole story. Do not build a behaviour plan on attachment theory alone. Watch the child, look at family context and SEND records, and ask for expert advice before you decide what support to put in place.

School-wide plans work best, not single teacher actions. Learners get confused if only one teacher uses attachment awareness (Bomber, 2007). Relational methods work if schools value wellbeing (Ford, 2017). Schools should see behaviour as communication (Perry & Szalavitz, 2006).

Attachment theory has limits, so use it carefully. Difficult behaviour isn't always about attachment. Learners from the same culture may differ from Western studies (Bowlby, 1969; Ainsworth, 1978). Use attachment insights as a guide, not a fixed plan.

Trauma shapes how learners cope in class. Ask CAMHS or educational psychology for expert support. Social work can help too (Perry, 2009; Bowlby, 1969; Fonagy, 2001). Keep these routes in mind.

‍

‍

‍

Building Safe Learning Environments

You open your classroom doors knowing that some children arrive already in crisis. Their nervous systems are primed for threat, their expectations shaped by experiences you may never know. Yet for six hours a day, you have the opportunity to offer something profoundly important: a consistent, attuned presence that tells them: you are safe, you belong here, your feelings matter, and I will be here tomorrow too.

This is not a small thing. For a child who has not experienced this elsewhere, it can be profoundly important. It may not erase their trauma or rewrite their history, but it provides a powerful counter-narrative.

It teaches their nervous system that safety is possible. It creates the foundation on which learning can stand.

Attachment theory gives us a language for understanding why a particular child struggles and a framework for how to help. But at its heart, attachment is about the profound human need for connection and the quiet power of a trusted adult who shows up, day after day, in the small moments that matter most.

‍

• Child Development Theories: For a broader understanding of child development, read our comprehensive guide to child development theories, which contextualises attachment within the wider field of developmental research.
• John Bowlby's Work: For a deeper dive into Bowlby's original framework, see our detailed article on Bowlby's attachment theory.
• Foundation for Secure Learning: Attachment builds on Erik Erikson's (1963) concept of trust versus mistrust in early development, the stage in which secure attachment is foundational.
• Maslow's Safety Needs: Understand how Maslow's hierarchy positions safety and belonging as prerequisites for learning (Maslow, 1943), because attachment is how those needs are met in practice.
• The Family System: Bronfenbrenner's (1979) ecological model reminds us that the child's family microsystem is the primary context for attachment formation.
• Unconditional Positive Regard: Carl Rogers' (1969) concept of unconditional positive regard aligns closely with secure attachment and the teacher's role in supporting emotional development.
• Nature Versus Nurture: Attachment is a key argument in the nature versus nurture debate, showing how early relational experiences shape development.
• Psychodynamic Roots: For historical context, attachment theory builds on Freud's (1923) psychodynamic thinking about early relationships, though with a focus on observable behaviour rather than unconscious drives.
• The Secure Base and Learning: Vygotsky's (1978) concept of the Zone of Proximal Development relies on a secure relationship with a more knowledgeable other, and attachment theory explains why that relationship carries so much weight.