Affectional Bonds—Bowlby on Inner Working Models and Expectation Fields

Share this Blog post

This is part III of my multi-part review of John Bowlby’s 1979 book The Making and Breaking of Affectional Bonds. In part I, I profiled Sir Richard Bowlby’s introduction to Affectional Bonds, which was added in 2005. In part II, I profiled criticisms of attachment theory that came from certain sectors of the feminist movement back in the 1960s and 70s. I also talked about Bowlby’s take on how attachment theory might affect women’s efforts to enter the workforce (again, back in the 1960s and 70s). In this part I’ll talk about Bowlby’s idea of Inner Working Models. If there is time I’ll also talk about how Inner Working Models may be viewed as expectation fields.

In Affectional Bonds, Bowlby makes the following statement:

Though [attachment theory] incorporates much psychoanalytic thinking, the theory differs from traditional psychoanalysis in adopting a number of principles that derive from the relatively new disciplines of ethology and control theory; by so doing it is enabled to dispense with concepts of psychic energy and drive and also to forge close links with cognitive psychology.

Essentially Bowlby is suggesting that we do away with vitalistic concepts such as psychic energy and drive, and embrace scientifically derived concepts such as innate behavioral systems (of which attachment is one) and Inner Working Models. But here’s where I have a beef with Bowlby. He “poo poos” a Freudian concept like psychic energy as being too vitalistic, but then calls us to embrace a concept drawn from cognitive science—Inner Working Models—without really telling us what an Inner Working Model is or what it does. (1) As an example consider this statement by Bowlby as he talks about what will be required as clinicians move from psychodynamic explanations to ones offered up by attachment theory. Bowlby states that clinicians of a psychodynamic bent will have to replace an

orally derived theory of internal objects by a theory of working models of the world and self that are conceived as being constructed by each individual as a result of his experience, that determine his expectations, and on the basis of which he plans.

Believe it or not, Bowlby is talking about Executive Function skills, of which working models and making plans are parts. Even though work on EF started in Britain back in the 1950s (thank you Wikipedia), EF as a discipline really did not come into its own until the 1970s and 80s. (I should mention that in the early days of its development EF shared much in common with control theory.) How was a clinician to move from “internal objects” to “Inner Working Models” when to do so would have required a herculean effort in terms of academic study. I hate to say it but Bowlby often talks about concepts (i.e., organismic biology, cybernetics, control theory, and Inner Working Models) as if these concepts are common knowledge requiring no further explanation. As I mentioned in part I, Bowlby constantly challenges you to gain familiarity in so many areas such as evolution, biology, ethology, cybernetics, control theory, spatial cognition, organic systems theory, not to mention the obvious areas such as developmental psychology, psychiatry, and Freudian psychodynamics. If I had to point to one reason Bowlby’s attachment theory never really caught on, I would point to the interdisciplinary breadth of knowledge required to fully understand the implications of attachment theory. So, in the rest of this blog I will try to demystify the topic of Inner Working Models.

Example of a topographic map

How can we begin to conceptualize what an Inner Working Model is and what it does? One way to get started is to imagine Inner Working Models as topographic maps. To the left (or above if you are reading by email) is a simple example of a topo map taken from the Internet. Topo maps give you a sense for the terrain, mainly how steep or gentle the slopes are. In the example here we see a series of contour lines. Contour lines represent “equal elevation.” See the major contour line labeled “700”? Everywhere along that line the elevation is 700 feet above some datum, usually sea level. When contour lines are spaced far apart (i.e., to the east in our example) the terrain is relatively gentle. When contour lines are closely spaced, the terrain is steep. If you were to walk along the red line from Equation Creek to the top of the hill shown (elevation 859 feet), you’d gain approximately 160 feet of elevation over a traverse of about a mile and a half. That’s a grade of about 2%. (For comparison, the steepest allowable highway grade in the US is 7%.) If you were to continue on that course heading down the backside of the hill (to the southeast) the grade drops to about 1.2%.

Topo maps represent the world to us the way it is, within the constraints imposed by cartographic practices (which get more precise all the time as we use more and more satellite imaging). Unlike our simple example, many topo maps contain symbols for things like railroads, highways, mines, houses, schools, churches, etc. And, yes, cognitive science shows that people (and many animals) build something like a topo map in their heads. We use these inner cognitive maps or models to navigate the environment. Honey bees will use inner cognitive maps to locate nectar and bring that nectar back to the hive. Migratory birds also use inner cognitive maps to make their yearly trek, often covering thousands of miles. There is a whole sub-discipline of cognitive science that studies what is known as spatial cognition—how we are able to know and navigate three dimensional space (see the reading list below). Architects and airline pilots tend to have well developed spatial cognition. (2) When Bowlby talks about how we build inner models based on our individual experiences, he is in large part talking about how our traverses across a particular terrain allow us to build inner cognitive models. Bowlby reveals:

There is now a mass of evidence [drawn from the world of spatial cognition] to support the view that exploratory activity is of great importance in its own right, enabling a person or animal to build up a coherent picture of environmental features which may at any time become of importance for survival.

Visually impaired persons have what is known as “blind sight.” A visually impaired person can walk around his or her home or even neighborhood and have very little difficulty. Why? They use their inner cognitive models to “see” by. Visioned persons do the same thing but they are not aware they are doing so (unless they’re making their way to the bathroom late at night in almost total darkness).

The main difference between topo maps and inner working cognitive models has to do with subjective experience, the way the world is individually experienced. Upon each “the way the world is” topo map in our heads, we impose our own “the way the world is experienced” topo map. As a simple example, a young adventurous person may view the steep traverse (shown by the red line in our example topo map) with excitement and anticipation. That excitement and anticipation is not represented on the example topo map. An elderly person who has mobility issues may view the same line with trepidation and even fear, again, neither of which emotional experience is represented on the map. What our topo map example does not show us are contour lines of “equal emotion” or “equal experience” if you will. One person may have a contour line of joy that runs parallel to the Equation Creek because in part they love to be near water. Another person may have a contour of dread, maybe even fear that runs along the Equation Creek because they had a bad experience with water as a young child. And, yes, our emotional topo map changes as we change. We become older and we no longer enjoy steep climbs. Or maybe we conquer our fear of water by learning how to swim.

So, it’s rather simplistic but Inner Working Cognitive Maps contain a layer that represents the world as it is, and it contains a layer that represents the world as we experience it. The two are inextricably intertwined. I think this is an important point. Often attachment types will say something like “when a child has a secure home base, he or she is free to explore the world.” Exploring does not pick up where the secure base leaves off. They work together. The inner cognitive map that is created through exploring the world contains the attachment experience, the attachment relationship. To move out into the unknown, the unmapped, is to know that one can move back to the known, the mapped, the safe base if need be.

Bowlby would regularly talk about what he called the “risk of risk.” He’s in essence talking about the two layers of Inner Working Models (and I’m sure there are other layers). The world “as it is” presents us with risk, maybe a really steep slope. The question then becomes If we take on that risk and we exceed our abilities, can we import help or aid into the system? That’s the risk of risk. Do we know where aid is located? Can we ask for aid? Can we accept aid? The risk of risk creates contour lines that simply do not appear on conventional topo maps. And in fairness to the Adult Attachment Interview (see note 1), maybe that’s what the AAI measures, the contours created by the risk of risk. As the AAI asks, “As a child, when you were hurt (e.g., when the challenges of the world exceeded your abilities) who did you turn to (e.g., who was available to provide aid) and what did you do (e.g., were you able to make your way back to that source of care and aid). (3)

So, I hope the above gives the reader a general sense for what inner cognitive maps are and how they are used. I’ve gone over my word count, so I’ll come back to the topic of expectation fields in part IV. But take note that in the above quote Bowlby states that Inner Working Models determine one’s expectations.


(1) – I ran into this same problem when I took the Adult Attachment Interview training. During the AAI training we were told that the AAI could be used to assess a person’s Inner Working Model with respect to attachment functioning. When I asked what this meant exactly, I received no real explanation. Again, the concept of Inner Working Models was presented as if common knowledge that required no further explanation.

(2) – In my December 8th, 2014 blog post entitled The Rise and Fall of Procedural Man (and Woman), I profiled Nicholas Carr’s 2014 book entitled The Glass Cage: Automation and Us. Carr’s book contains an interesting sub theme: The current rise of automation may mark the fall of Procedural Man. It’s beyond the scope of this note but suffice it to say that there is a close relationship between Inner Working Models and what are known as procedural memories. When you ride a bike, perform surgery, or fly an airplane you are using procedural memories. Carr bemoans the fact that as automation rises, our ability to develop procedural memories falls. As Carr tells us, architects no longer design, airline pilots no longer fly, and surgeons no longer operate because all increasingly depend heavily on computer systems to do the so-called heavy lifting. Carr suggests that as we become dependent on automated systems we are no longer able to detect the warning signs of impending doom, as evidenced by the recent rash of high-profile airline crashes (i.e., the July 6, 2013, crash of Asiana Flight 214 as it approached San Francisco). All of this points out how important development and maintenance of robust Inner Working Models and procedural memories are to certain professions: surgeon, airline pilot, architect, cruise ship captain, nurse, dentist, crane operator, even over-the-road truckers. Some futurists argue that allowing automated systems to do the heavy lifting is a good thing, and mental models will go the way of the buggy whip. As an example, Google is hard at work developing the self-driving car, a car that will allow us to navigate the environment without having to utilize any maps at all whether real or mental. Bowlby’s theory of attachment challenges us to develop robust Inner Working Models. However, Inner Working Models may be “so last century” as the kids say.

(3) – Parenting can be a tricky business. Parents have to allow their kids to experience some level of risk (the “risk”) while keeping them safe (the “risk of risk”). Learning how to ride a bike or to swim come to mind. In these two situations the two layers stay fairly close together—the risk, and the risk of risk. On occasion, the two layers are ripped apart one from the other. Imagine you are a parent or guardian and you see your charge heading for the street. You frantically run to the curb and grab her up. What are your emotions? Throttle him for putting himself in such potential danger, and hug her for dear life. You as a parent are going in two different directions at the same time, along two different layers. It’s at times like these that we can experience the two layers as separate. Mourning would be another such experience: to grab on to the dearly departed for dear life, and to let them go. These “two layer” experiences do not come that often, but when they do, they are profound. Maybe in the “space between” (“between and betwixt” as Father Richard Rohr used to say) we experience something approaching the divine. Here are a few other places where we may experience “two layers”: dance, comedy, certain rituals, athletics, and religion.

Spatial Cognition, Cognitive Mapping and Mental Models Reading List:

Downs, R. & Stea, D. (Eds.). (1973). Image and environment—Cognitive mapping and spatial behavior. New Brunswick, NJ: Transaction Publishers.

Esser, A. (Ed.). (1971). Behavior and environment—The use of space by animals and men. Proceedings of an International Symposium held at the 1968 meeting of the American Association for the Advancement of Science in Dallas, TX. New York: Plenum Press.

Fauconnier, G. (1994). Mental spaces: Aspects of meaning construction in natural languages. Cambridge: Cambridge University Press.

Fauconnier, G. and Turner, M. (2002). The way we think—Conceptual blending and the mind’s hidden complexities. New York: Basic Books.

Golledge, R. (Ed.). (1999). Wayfinding behavior—Cognitive mapping and other spatial processes. Baltimore, MD: The John Hopkins University Press.

Hutchins, E. (1995). Cognition in the wild. Cambridge, MA: MIT Press.

Kitchin, R. & Blades, M. (2002). The cognition of geographic space. London: I.B. Tauris.

Lakoff, G. and Johnson, M. (1999). Philosophy in the flesh—The embodied mind and its challenge to western thought. New York: Basic Books.

Laszlo, E., Artigiani, R., Combs, A. and Csányi, V. (1996). Changing visions—Human cognitive maps: past, present, and future. Westport, CT: Praeger.

Nisbett, R. (2003). The geography of thought—How Asians and Westerners think differently…and why. New York: Free Press.