Seeing and (Not) Believing in Anorexia

What can we learn from case studies of anorexia in the blind?

Posted Dec 16, 2012

Do we need to be able to see to be dissatisfied with our bodies? Could blindness reduce the likelihood of developing an eating disorder that involves negative body image? Or might not being able to see make us all the more likely to develop body-image distortions that go unchecked by vision? Congenital blindness and anorexia don’t often seem to coexist, but there are a number of published case studies that explore how vision and its absence, body shape and size, and psychological pathology can interact. Here I’ll outline one of the richer and more detailed accounts, which is both interesting in its own right and a potential source of insights into anorexia in sighted people.

Thomas et al. (2012) describe the case of ‘Ms A’, who described herself as having ‘3% of full vision’: she could detect only high contrast, some colour, and large shapes under optimal conditions, but otherwise nothing at all. Her anorexia began at age nine, with obsessive-compulsive disorder predating its onset, and anorexia and depression being formally diagnosed at age 14. When she was first admitted to a residential treatment programme, aged 19, Ms A denied the seriousness of her low bodyweight (BMI 15.9) and didn’t manifest the psychological criterion for anorexia, ‘intense fear of gaining weight or becoming fat’. Instead, she said that she refused food in order to provide a sense of ‘containment, safety, and control’ through the practice of self-restriction, as well as to elicit maternal behaviours from her caregivers: she said she had ‘an overwhelming desire for the experience of childishness and care’.

It wasn’t till nearly a year later that Ms A began to report concerns relating to body image. She described feeling newly competitive, primarily at meal times, with a sighted female relative who also had anorexia, assessing the relative’s dramatic weight loss with her limited vision and when hugging her. She also seemed newly interested in how her body might look to potential romantic partners, declaring she was too ‘flat-chested’ – not a typical anorexic value judgement, but one that testifies to an awareness of how the female body is culturally idealised. By this point, Ms A also seemed less concerned with being cared for, and more determined to become independent – a goal which she was aware was in conflict with her dietary restriction. Other possible reasons for her newly reported body-image concerns are the possibility (sadly) of social contagion during the residential phase of her treatment, and conversely the development of new insight or willingness to talk about things, since by this time the interpersonal side of things had made her more motivated for change.

Talking about her disturbed body image, Ms A described typical features of anorexia in the sighted, including checking behaviours, body avoidance (e.g. avoiding tight clothing), and comparisons with others. As she put it, ‘I do all of the checking, but without the eyes’. She, like other blind sufferers from anorexia, used tactile body checking – feeling for bony protrusions in face, ribcage, spine, and hips, and assessing how tight her clothes felt.  She would also try to engage in the ‘mirror checking’ that’s so common in sighted anorexics: she described how she would stand very close to full-length mirrors to use the patterns of light and dark to try to assess her shape. ‘I like to pretend I can see a form’, she said. Usually, because she could only see ‘a blob’, this kind of checking would simply confirm to her that her body was too large. When comparing herself with others, Ms A used changes in air pressure to estimate people’s body size, and auditory cues like voice location and pitch to gauge height and weight respectively – a practice common among the blind.

As in the treatment of sighted people, Ms A’s treatment involved self-monitoring, which helped identify and reduce superfluous movements in activities like dressing and showering: she became able to stop feeling her hipbones while dressing or her arms while showering, for example. Although she was adamant that the voice and air-pressure techniques did allow her to assess quite accurately the appearance of other people, she did also admit that with other people she ‘probably imagines the best of all possible worlds’ – for example, ‘a slim figure and attractive proportions’. To help with this, she was asked to imagine less idealised versions of appearances that would be equally consistent with the data she collected. As for the mirror checking, she was asked to describe only what she could actually see within her 3% vision. She then compared the very limited information actually available to her – ‘I can’t see my body because I’m wearing dark clothes… I can’t detect any contour… I’m used to taking a lot more from it’ – with all the things she’d usually conclude, for example that she had ‘large hips, large thighs, and strange coloring’. This kind of assessment of how the reality doesn’t warrant the conclusions drawn from it is key to dismantling all sorts of cognitive distortions in anorexia generally.

Both the similarities and the differences in how anorexia is manifested and treated in sighted people and in this blind person raise some interesting questions. The parallels in body checking are very striking, and the complementing of visual input by means of tactile input is certainly not uncommon in anorexia in general. Body checking is one of the major ways in which negative body image is maintained in eating disorders. The danger of checking behaviours, whether visual or tactile, isn’t just that they maintain an obsessive focus on body shape and size, but more specifically that by definition the parts of the body that are ‘checked’ the most often are those the sufferer is least ‘satisfied’ with, or has most problematic associations with, and therefore the practice of the behaviour is the most likely to result in the confirming perception of a ‘flaw’. This is in turn usually amplified to result in a more negative evaluation of the whole body, which in turn feeds into a more total negative self-evaluation, which then requires more checking to ‘make sure things aren’t getting worse’, or something to that effect. And thus the vicious circle perpetuates itself.

An interesting additional question here is whether limited sensory feedback of the kind a blind sufferer is restricted to promotes or diminishes this selective attention to perceived flaws, or indeed whether the different types of input (tactile versus visual) have no effect. On the one hand, one might think that the relative lack of actual sensory feedback could give the cognitive distortions freer rein to impose impossible demands without corrective visual feedback. On this view, already extreme states of emaciation would be readily responded to as simply not extreme enough; there’d be no correction through the shock of seeing the whole starved body and realising how horrific it looks, for example. But the apparently relatively low incidence of anorexia in the blind would seem to speak against that conclusion (though underreporting may be an issue here).

Maybe, conversely, then, the lack of visual input provides a kind of protection against those cognitive distortions. Certainly representations of idealised thinness in the visual media wouldn’t be the ever-present reminder of one’s own ‘flaws’ that sighted people have to deal with. And while it is sometimes said that in depression everything looks grey, maybe being limited primarily to touch might reduce the ability of distorting cognitive patterns to acquire apparent confirmation from the checking behaviours. But while vision is cognitively penetrable – i.e. it can be influenced by beliefs, expectations, mood, etc. – the same is true, perhaps to an even greater extent, of touch, not least because active exploratory movements are required in touch but not in vision, where rapid eye movements (saccades) are automatic, and we can’t avoid seeing, except by closing our eyes. And to judge from Ms A’s testimony, it does sound as though the feedback loops just as easily get just as pernicious as with vision. Even if there is an effect of lack of sight in making onset of anorexia slightly less likely, it certainly wouldn’t seem to provide any real protection once the eating disorder has begun.

Vision can give the impression of taking in a scene or an object – for example, a whole body – easily in one go, whereas touch is more dependent on inference from the smaller surface region of an object currently being touched, and has to proceed more slowly to cover the rest of a surface – working up the arm, say, to feel where and how much the elbow joint protrudes and the upper arm then becomes concave in shape. This would seem to mean there would be a dramatic difference in how vision and touch function as components of anorexic checking behaviours. But in practice, the way vision functions in this context is often through exclusive focus on a single body part, resulting in similar limitations as in touch. And, as noted, visual checking is often also combined with tactile checking: pressing in the tummy or inner thighs and looking in the mirror at what happens, for example. This might mean that in anorexia vision and touch lend themselves equally readily to involvement in the kinds of cognitive errors that are characteristic of anorexia: over-generalisation (this part of my thigh looks/feels big, therefore I’m fat), magnification (quite literally magnifying the size of the full tummy after eating, visually or through touch), discounting the positive (ignoring physical features one might find pleasing to focus on supposedly less pleasing ones), and all-or-nothing thinking (either I can see/feel every rib clearly or I’m fat).

The more restricted input of too narrowly focused vision or of touch alone might thus help maintain the eating disorder in rather similar ways, once those errors have kicked in because of loss of cognitive flexibility due to starvation and associated depression. The same principles apply not just to self-checking but to comparisons with others. In both visual and tactile checking and comparing behaviours, what in normal circumstances is a useful adaptation – the ability to draw inferences about oneself and others from limited sensory information – can easily become detrimental if the cognitive context changes even slightly, as is bound to happen when the physiological context changes due to semi-starvation.

Just as the blind person has to learn formally to extrapolate from limited tactile input, so all sighted people learn quickly and automatically to extrapolate from limited visual input. The gappy nature of visual perception is well documented, for example in studies on phenomena like change blindness. We don’t fill in all the gaps when we see – we often just jump to a conclusion like ‘more of the same here’. Indeed, the philosopher Dan Dennett (Consciousness Explained, p. 354) has argued that ‘jumping to conclusions’ is a fundamental part of how vision works. Although you have clear vision only in the very small foveal area in the centre of the eye, if you go into a room wall-papered with, say, identical portraits of Marilyn Monroe, what do you see? You don’t see, as the limitations of saccades and foveal fixation would suggest, one clear portrait and lots of bleary blobs, before turning to the next, foveating that, identifying it as Marilyn, and so on. You see at a glance, ‘instantly’, that they are all the same – and would notice straight away if one had a hat or a silly moustache. You see ‘all Marilyns’. You jump to the conclusion that they’re ‘all Marilyns’, and because you have visual mechanisms like pop-out detectors to alert you to striking differences, most of the time you’re probably right. You aren’t right, however, when instead of jumping to the conclusion ‘all Marilyns’ on the basis of one or two Marilyns you jump to the conclusion ‘all fat’ on the basis of a less than rock-hard thigh.

There’s some evidence that body dysmorphic disorder is associated with changes specifically in visual processing, specifically reduced activation in brain areas responsible for holistic processing or a scene. The trouble is that it’s easy to forget that vision works by means of shortcuts and is therefore fallible in all kinds of ways. It’s common to attribute to vision a kind of objectivity that we don’t attribute so much to the other senses. The metaphorical association of seeing with knowing is deeply embedded in everyday language – think of metaphors like ‘I see where you’re coming from’, ‘she gave a very clear overview of the topic’, and so on. This may be due partly to the highly detailed kind of access it gives us to the world, and to its capacity to take in a great deal seemingly at once. But vision is of course as fallible as all the other senses, and its fallibilities can become dangerous when other aspects of cognition are distorted by, for example, very low body weight.

Cognitive behavioural therapy can very efficiently draw the sufferer’s attention to these kinds of biases, and offer methods for correcting for them. The starting point, as always in recovery from eating disorders, is acknowledging that bodily changes result in cognitive ones. It’s important to remember that this applies just as much to vision as to touch, and that there are all sorts of reasons why in anorexia, seeing shouldn’t be believing.



With thanks to Miriam for drawing my attention to some case studies on anorexia and blindness.