The Map We Carry

How Attachment Experience Becomes the Working Model

There is a moment, somewhere in the accumulation of early relational experience, when the feedback loop stops being just a loop and becomes something more durable. When the repeated experience of bids sent and answered, or sent and missed, stops being a series of individual events and becomes a set of expectations. A map. An internal working model of how relationships work, what other people can be counted on for, and whether the self is the kind of person whose needs are worth meeting (Bowlby, 1982; Main et al., 1985; Waters & Waters, 2006).

John Bowlby called it the internal working model, and he placed it at the center of his understanding of how early attachment shapes the rest of a life (Bowlby, 1982). The working model is not a conscious belief system. It is not something a person decides to hold. It forms below the level of awareness, out of the accumulated evidence of early relational experience, and it travels with the person into every relationship they will ever have, shaping what they expect, what they reach for, what they brace against, and what they allow themselves to need (Bowlby, 1982; Collins & Read, 1990; Main et al., 1985; Mikulincer & Shaver, 2016).

Understanding the working model is essential to understanding attachment across neurologies, because the working model is where the differences between autistic and non-autistic developmental pathways become most consequential for adult life. What gets built in infancy and childhood does not stay in infancy and childhood. It arrives, sometimes decades later, in the room where two people are trying to build something together, and it shapes what happens there in ways neither person may fully understand (Fraley, 2002; Hazan & Shaver, 1987; Mikulincer & Shaver, 2016).

How the Working Model Forms

The working model is built from the same material that builds trust: the feedback loop between the person and their relational environment. Every time a bid for connection is sent and accurately received, the working model registers a data point in the direction of safety. Every time a bid goes unanswered, or is met with a response that misses the mark, the working model registers a data point in the direction of wariness. Over thousands of interactions, those data points accumulate into something that functions like a prediction: this is what I can expect from the people around me, this is what relationships do, this is who I am in relation to others (Ainsworth et al., 1978; Bowlby, 1982; Main et al., 1985; Waters & Waters, 2006).

The working model is always an adaptation. It is the developing nervous system's best attempt to make sense of the relational environment it has encountered and to prepare itself for more of the same (Bowlby, 1982; Main et al., 1985). A child whose early environment was reliably responsive builds a working model oriented toward trust, one that expects bids to be received, that allows vulnerability because vulnerability has proven safe, and that can tolerate the inevitable moments of misattunement without interpreting them as evidence that connection is fundamentally unavailable (Ainsworth et al., 1978; Bowlby, 1982; Waters & Waters, 2006). A child whose early environment was inconsistent, unresponsive, or accurately calibrated in the wrong direction builds a working model oriented toward something more self-protective, one that expects less, reaches more cautiously, and interprets ambiguous relational signals through a lens of wariness rather than confidence (Ainsworth et al., 1978; Bowlby, 1982; Mikulincer & Shaver, 2016).

Neither working model is chosen. Both are reasonable. Both are the nervous system doing exactly what it is supposed to do: learning from experience and preparing for the future on the basis of what the past has taught (Bowlby, 1982; Fraley, 2002; Main et al., 1985).

The Working Model in Non-Autistic Development

For people whose development follows the non-autistic pathway, the working model is built primarily through the continuous embodied feedback loop that characterizes non-autistic social development from the beginning (Ainsworth et al., 1978; Bowlby, 1982; Main et al., 1985). Body empathy, as defined by MacMillan, is the immediate, felt reception of social information from the faces and bodies of others. It means that the non-autistic person is receiving relational data constantly, registering in real time how others are responding to them, and updating their working model accordingly. The process is largely automatic, largely below conscious awareness, and largely continuous.

This has a particular implication for the working model in non-autistic development: it is highly responsive to the social environment. Because the non-autistic nervous system is continuously receiving and integrating social feedback, the working model is being updated, tested, and revised on an ongoing basis (Bowlby, 1982; Collins & Read, 1990; Fraley, 2002). This is a strength: it means the non-autistic working model has a built-in capacity for revision, a responsiveness to new relational experience that allows the map to be redrawn when the territory changes (Fraley, 2002; Mikulincer & Shaver, 2016).

But it is also a vulnerability. A working model that is highly responsive to the social environment is also highly susceptible to the quality of that environment. Non-autistic people whose relational environments consistently offer distorted feedback, whose working models have been built from experience of inconsistency, emotional unavailability, or chronic misattunement, carry maps that are oriented toward wariness with the same automatic fluency that securely attached non-autistic people carry maps oriented toward trust (Ainsworth et al., 1978; Bowlby, 1982; Mikulincer & Shaver, 2016). The responsiveness of the system cuts in both directions.

The Working Model in Autistic Development

For people whose development follows the autistic pathway, the working model is built through a different process and carries a different set of strengths and vulnerabilities.

Because the autistic developmental pathway does not run on the same continuous stream of embodied social feedback, the working model does not update in the same automatic, ongoing way. Autistic people build their understanding of the relational world more sequentially, more deliberately, and more through explicit experience than through ambient social osmosis (Jaswal & Akhtar, 2019; Milton, 2012). This is a strength: when clear, accurate, and consistently calibrated relational experience is available, the autistic working model can be built on genuinely solid ground. Yet it also means that when that experience is not available, the working model may be built on whatever is accessible, which is often the distorted feedback that autistic people receive from environments that misread their social presentation (Bottema-Beutel et al., 2021; Sasson et al., 2017).

The particular vulnerability of the autistic working model is this: because it is built more deliberately and updated less automatically, it may be less immediately responsive to new relational experience than the non-autistic working model. A pattern of expectation, once established, may be harder to revise through the kind of ambient social recalibration that non-autistic people experience more or less continuously. This does not mean the autistic working model cannot be revised. It can, and it is, across the lifespan (Fraley, 2002; Mikulincer & Shaver, 2016). But revision may require more intentional conditions, clearer and more explicit relational experience, and more time than the non-autistic pathway typically needs (Cage et al., 2018; Crompton et al., 2022; Huang et al., 2022).

There is also something that accumulates specifically in the autistic working model that has no direct parallel in non-autistic development: the experience of having been consistently misread. When the relational environment has repeatedly failed to receive autistic attachment bids accurately, not through malice but through neurotype mismatch, the working model builds in a particular direction (Milton, 2012; Sasson et al., 2017). It learns not only that connection is uncertain but that the self's way of reaching for connection is somehow wrong, that the channels through which the autistic person naturally expresses need are not the channels the world is equipped to receive (Bottema-Beutel et al., 2021; Jaswal & Akhtar, 2019; Pearson & Rose, 2021). That is a more specific and more isolating lesson than simple relational wariness, and it shapes the working model in ways that matter for every adult relationship that follows (Bradley et al., 2021; Cage & Troxell-Whitman, 2019; Evans et al., 2024).

What the Working Model Does in Adult Relationships

By the time a person enters adult relationships, the working model is well established. It is not visible. It does not announce itself. But it is present in every significant relational encounter, shaping perception, expectation, and response in ways the person may not be able to fully articulate even when they can sense its effects (Collins & Read, 1990; Hazan & Shaver, 1987; Mikulincer & Shaver, 2016).

A person whose working model is oriented toward trust enters a new relationship expecting, at some level, that bids will be received, that vulnerability will be met with care, and that the inevitable moments of misattunement are repairable rather than catastrophic (Bowlby, 1982; Hazan & Shaver, 1987; Mikulincer & Shaver, 2016). They can tolerate uncertainty in a relationship without immediately interpreting it as abandonment. They can repair conflict without experiencing it as evidence that the relationship is fundamentally broken. Their map tells them that the relational territory is navigable, and they move through it accordingly.

A person whose working model is oriented toward wariness enters the same relationship with a different map. They may expect, at some level, that bids will be missed. That vulnerability will be used against them or will simply go unregistered. That conflict signals something more serious and less repairable than it actually is. Their map tells them that the relational territory is less safe than it appears, and they navigate it with the care that map demands, which can look, from the outside, like distance, defensiveness, or difficulty with intimacy, even when what is actually happening is the reasonable caution of someone whose early experience taught them that caution was warranted (Bowlby, 1982; Collins & Read, 1990; Mikulincer & Shaver, 2016).

For autistic adults, the working model carries an additional layer of complexity. The wariness that has built up through years of misread bids and conditional acceptance is not simply attachment insecurity in the classic sense. It is also a learned uncertainty about whether the self's way of connecting is legible to others at all (Milton, 2012; Pearson & Rose, 2021; Sasson et al., 2017). The autistic adult may want closeness as much as anyone, may be reaching for it with genuine intention, and may simultaneously be bracing for the familiar experience of reaching and not being received (Bradley et al., 2021; Cage & Troxell-Whitman, 2019; Jaswal & Akhtar, 2019). Both things are true at once, and navigating that tension in adult relationships requires more than goodwill from the people on both sides of it.

The Map Can Be Revised

This is the part that matters most, and it is where Bowlby's framework is most hopeful. The working model is not fixed. It is built from experience, which means it can be rebuilt by experience (Bowlby, 1982; Fraley, 2002; Main et al., 1985). Adult relationships, when they are genuinely responsive and accurately calibrated to the person they are in relationship with, offer the nervous system new evidence (Hazan & Shaver, 1987; Mikulincer & Shaver, 2016). Evidence that contradicts the map. Evidence that bids can be received, that vulnerability can be safe, that the self's way of connecting is legible and welcome.

That revision does not happen quickly or easily, particularly when the working model has been built from years of consistent experience pointing in the other direction (Fraley, 2002; Mikulincer & Shaver, 2016). And for the autistic working model specifically, revision may require more than the ambient social recalibration that non-autistic development relies on (Cage et al., 2018; Crompton et al., 2022; Huang et al., 2022). It may require relationships that are explicit about what they are offering, that name the safety they are providing rather than assuming it will be felt automatically, and that are patient with the horizontal surfaces of the autistic staircase, the periods of consolidation and testing, that precede the next step of upward growth.

Researchers who have asked autistic adults directly what has helped their development across the lifespan find the same answer consistently: people. Specific people who understood, accepted, and expected something of the autistic person at the same time (Cage et al., 2018; Crompton et al., 2022; Huang et al., 2022). Not programs or interventions in the abstract, but the experience of being held by someone who saw them accurately and did not require them to change their shape to be welcomed (Cage et al., 2018; Pearson & Rose, 2021). That experience, when it comes, begins to revise the working model. It adds new data points. It offers the nervous system evidence it has not had before.

And the nervous system, autistic or non-autistic, does what it has always done with new evidence: it learns. It updates the map. It begins, slowly and on good grounds, to expect something different from the relational world than what the early evidence suggested (Bowlby, 1982; Fraley, 2002; Mikulincer & Shaver, 2016).

That is what Post 4 is about: what happens when two people, each carrying their own working model, each shaped by their own developmental pathway, come into close relationship with each other, and what it takes to build something that works for both.

This is Series II — Attachment Across Neurologies: How Trust, Security, and Relationship Maps Form Differently

This was post 3.

1. The First Thing We Need: How Trust Forms Across Neurologies
2. The Shape of Security: What Attachment Looks Like Across Neurologies
3. The Map We Carry: How Attachment Experience Becomes the Working Model
4. When Two Maps Meet: Attachment in Neurodiverse Relationships



References

Ainsworth, M. D. S., Blehar, M. C., Waters, E., & Wall, S. N. (1978). Patterns of attachment: A psychological study of the Strange Situation. Lawrence Erlbaum Associates.

Bottema-Beutel, K., Kapp, S. K., Lester, J. N., Sasson, N. J., & Hand, B. N. (2021). Avoiding ableist language: Suggestions for autism researchers. Autism in Adulthood, 3(1), 18–29. https://doi.org/10.1089/aut.2020.0014

Bowlby, J. (1982). Attachment and loss: Vol. 1. Attachment (2nd ed.). Basic Books. Original work published 1969.

Bradley, L., Shaw, R., Baron-Cohen, S., & Cassidy, S. (2021). Autistic adults’ experiences of camouflaging and its perceived impact on mental health. Autism in Adulthood, 3(4), 320–329. https://doi.org/10.1089/aut.2020.0071

Bretherton, I. (1992). The origins of attachment theory: John Bowlby and Mary Ainsworth. Developmental Psychology, 28(5), 759–775. https://doi.org/10.1037/0012-1649.28.5.759

Cage, E., Di Monaco, J., & Newell, V. (2018). Experiences of autism acceptance and mental health in autistic adults. Journal of Autism and Developmental Disorders, 48(2), 473–484. https://doi.org/10.1007/s10803-017-3342-7

Cage, E., & Troxell-Whitman, Z. (2019). Understanding the reasons, contexts and costs of camouflaging for autistic adults. Journal of Autism and Developmental Disorders, 49(5), 1899–1911. https://doi.org/10.1007/s10803-018-03878-x

Cassidy, J., & Shaver, P. R. (Eds.). (2016). Handbook of attachment: Theory, research, and clinical applications (3rd ed.). Guilford Press.

Collins, N. L., & Read, S. J. (1990). Adult attachment, working models, and relationship quality in dating couples. Journal of Personality and Social Psychology, 58(4), 644–663. https://doi.org/10.1037/0022-3514.58.4.644

Crompton, C. J., Hallett, S., McAuliffe, C., Stanfield, A. C., & Fletcher-Watson, S. (2022). “A group of fellow travellers who understand”: Interviews with autistic people about post-diagnostic peer support in adulthood. Frontiers in Psychology, 13, Article 831628. https://doi.org/10.3389/fpsyg.2022.831628

Crompton, C. J., Ropar, D., Evans-Williams, C. V. M., Flynn, E. G., & Fletcher-Watson, S. (2020). Autistic peer-to-peer information transfer is highly effective. Autism, 24(7), 1704–1712. https://doi.org/10.1177/1362361320919286

Evans, J. A., Krumrei-Mancuso, E. J., & Rouse, S. V. (2024). Autistic masking in relation to mental health, interpersonal trauma, authenticity, and self-esteem. Autism in Adulthood, 6(2), 229–240. https://doi.org/10.1089/aut.2023.0051

Fraley, R. C. (2002). Attachment stability from infancy to adulthood: Meta-analysis and dynamic modeling of developmental mechanisms. Personality and Social Psychology Review, 6(2), 123–151. https://doi.org/10.1207/S15327957PSPR0602_03

Hazan, C., & Shaver, P. (1987). Romantic love conceptualized as an attachment process. Journal of Personality and Social Psychology, 52(3), 511–524. https://doi.org/10.1037/0022-3514.52.3.511

Howlin, P. (2021). Adults with autism: Changes in understanding since DSM-III. Journal of Autism and Developmental Disorders, 51(12), 4291–4308. https://doi.org/10.1007/s10803-020-04847-z

Huang, Y., Arnold, S. R. C., Foley, K.-R., & Trollor, J. N. (2022). A qualitative study of adults’ and support persons’ experiences of support after autism diagnosis. Journal of Autism and Developmental Disorders, 52(4), 1634–1651. https://doi.org/10.1007/s10803-021-05032-0

Jaswal, V. K., & Akhtar, N. (2019). Being versus appearing socially uninterested: Challenging assumptions about social motivation in autism. Behavioral and Brain Sciences, 42, Article e82. https://doi.org/10.1017/S0140525X18001826

Main, M., Kaplan, N., & Cassidy, J. (1985). Security in infancy, childhood, and adulthood: A move to the level of representation. Monographs of the Society for Research in Child Development, 50(1/2), 66–104. https://doi.org/10.2307/3333827

Mikulincer, M., & Shaver, P. R. (2016). Attachment in adulthood: Structure, dynamics, and change (2nd ed.). Guilford Press.

Milton, D. E. M. (2012). On the ontological status of autism: The “double empathy problem.” Disability & Society, 27(6), 883–887. https://doi.org/10.1080/09687599.2012.710008

Pearson, A., & Rose, K. (2021). A conceptual analysis of autistic masking: Understanding the narrative of stigma and the illusion of choice. Autism in Adulthood, 3(1), 52–60. https://doi.org/10.1089/aut.2020.0043

Sasson, N. J., Faso, D. J., Nugent, J., Lovell, S., Kennedy, D. P., & Grossman, R. B. (2017). Neurotypical peers are less willing to interact with those with autism based on thin slice judgments. Scientific Reports, 7, Article 40700. https://doi.org/10.1038/srep40700

Waters, H. S., & Waters, E. (2006). The attachment working models concept: Among other things, we build script-like representations of secure base experiences. Attachment & Human Development, 8(3), 185–197. https://doi.org/10.1080/14616730600856016