2. Where Difference Begins

In an interview, Dominik,² a 40-year-old father of two, and a migrant from Poland, now living in Munich, Germany, tells us:³

When I see a transvestite, or transsexual person, I don’t know, a so-called woman with a beard, this is not a sight that is shocking to me. Sometimes I meet a woman, she is very well-made up but I can see that she has an Adam’s apple.

Dominik alludes to the Austrian artist known as Conchita Wurst, who gained international popularity after winning the Eurovision Song Contest in 2014. She established herself as a LGBTQ+ icon. On stage, she often appeared in feminine clothes, with long hair, and a beard. Describing the presence of queer people on the streets in Munich, Dominik tries to “detect” what he perceives to be a person’s “true gender” and detach it from what he calls “their performance” and suggests he can indeed establish it just by looking.

Dominik does not personally know any “transsexual people, or any transvestites”, as he says. But seeing a person, he projects his imagination of masculinity and femininity, gained through socialisation in Poland, onto them. Bech’s (2014) notion of “surface” is useful to analyse the narratives such as those produced by Dominik. “Surface” expresses how masculinities and femininities are constructed in the process of looking at others—strangers—in urban public spaces. As urban encounters are superficial and fleeting, it is the “surface” of the other—aestheticised and sexualised, ethnicised and racialised—which becomes the subject of evaluation. The observer seems to stabilise the observed through their gaze and looks at them analytically as though this person would exist in some accessible state available for comparison and judgement (Crary 1999: 300).

Anita, a Polish born Londoner, an entry-level office worker and social mother to her partner’s biological son,⁴ gives an example of what she considers to be “typically British femininity”:

Recently, I just didn’t know what to do anymore. A woman came [to school] wearing a pink tracksuit. With that topknot on her head, flip-flops, and a sign spelling ‘VIP girl’, right? I am thinking: Woman, you have a child, you are not a girl anymore, you are a woman. Sometimes I really feel the itch to come up to them and [she makes a strong shaking motion], fuck, get a grip, really. (Lisiak 2017: 47)

Anita’s narration reveals her disgust turning into anger in the situation that seems irreconcilable with the norms she knows from home, and the class, ethnic, and gender prejudice she harbours. But this quotation is interesting mostly because it demonstrates that just by looking at another woman, Anita thinks she can establish who she is: here, English, white and working class. Her subjective optical impression becomes a kind of ‘objective, shared truth’.

Both Dominik and Anita are fixated on perceived visual differences between themselves and the others, and between the ‘norm’ and the ‘aberration’. As looking at others is relational, the gendered or ethnicised difference is constructed in relation to other white men and women. In all interviews, when Polish women talk about people of colour or Muslims, they always mention their ethnicity or religion (Lisiak 2017: 50); when men speak of other men and women’s sexuality that is non-heteronormative, they refer implicitly to white Germans or white English people. Conversely, Muslim men are denied queer sexuality (Wojnicka and Nowicka 2021).

These narrations are instructive for at least two intertwined reasons. First, when projecting their ideas around gender, sexuality, ethnicity, and race on the bodies of others, the interviewees categorise others as members of some groups, such as German, British, straight or queer, men or women. They do so by placing some selected features of other people’s appearance and performance in focus and by ignoring others. Second, attending to some and ignoring other differences is not accidental. Attending to difference and ignoring difference reflects and enacts power. As Wojnicka and Nowicka (2021) and Lisiak (2017) explicate, Polish migrants in England and in Germany do not hesitate to demonstrate dismissive and discriminatory opinions of Muslim men and women as the racialisation and stigmatisation of Muslim bodies is largely legitimised in public discourses in these countries, as well as in Poland, which remains an important point of reference for these migrants. For Polish women in England and Germany, depicted by Lisiak (2017), white female bodies are visible, as they are also valuable, while non-white and Muslim bodies are invisible and misrecognised. What is also invisible, in the sense of not being addressed consciously, is the bodies’ whiteness (Ahmed 2007), for whiteness is the norm in these countries. Aligning with the first and devaluating the second, Polish women signal their belonging to the white hegemonic group. Those who, like Anita, stress their distance to British women, try to overcome their experience of downward class mobility by contrasting their education and civility to the perceived lack of it amongst lower-class British women who inhabit the same neighbourhoods in London or Birmingham and where the Polish migrant women do not feel they really belong.

Similarly, Polish men in Germany described by Wojnicka and Nowicka (2021) establish a proximity to German men and distance themselves from Muslim men if they read them as heterosexual. But they also distance themselves from German men by questioning their “masculine” features: in the narrations we collected, German masculinity is presented as “harmless” or even “too civilised”, tamed, having some “feminine” features. German men are portrayed as paying exaggerated attention to their looks and avoiding physical confrontation. Furthermore, they are seen by Polish men as not chivalrous enough, and thus lacking respect towards women and vulnerable people. Polish masculinity emerges here as a golden middle between an “aggressive Muslim” and “feminised German” male other.

Barbara, a Polish born middle-aged mother of two who lives in a mid-sized town in the Midlands, UK, interviewed in 2011⁵ tells me about the relations between Polish migrants in England and other nationals. She gives examples of some minor tensions between English and Polish people and then says:

I know such stories that they [English] don’t like us [Polish] but I think if that many different nationalities would come to Poland, like it is in England, I don’t think it would go well. Polish people don’t like Kurdish, Bangladeshi, as they [Poles] say ciapaty, or how else they call them? Dirty, ciapaci. That’s what they say.

She distances herself from this term and adds quickly “for me all are humans” but later in the interview she mentions that some of her Polish (former) friends who turned her down because she “married a ciapaty” (cf. Fiałkowska 2018), and she accepts this term as describing her own husband.

Another Polish woman, Anna, interviewed in 2011 in the same town, manager in a shop, in her twenties, recalls her first days in England and how she tried to find accommodation. She replied to an ad in a newspaper and went, accompanied by her friend, to meet the potential landlord:

There came a guy, just imagine, with a car, ciapaty and [said to us] come here my girls, come quickly, get into the car, get in! And I say no, our luggage is here. And he continues, get in and takes my friend’s hand, tries to draw her into his car… I got scared, he could kidnap us.

She goes on describing how she and her friend could free themselves from the man and how she feared him. I asked her to explain what she means by ‘ciapaty’ and she explains: “you know, all kinds of nationalities, all Indians, I don’t know”. A similar explanation was given to me in an interview in 2011 by Bianka and her partner, both in their mid-twenties. She works as a picker and packer in a large warehouse in the Midlands; he is a truck driver. Both say, “We have no negative feelings towards… Indians, ciapaci” and Bianka clarifies for me:

In general, in England, I think I’ve heard it for the first time in Coventry, this is a term describing a person who is not black and not white, who is in-between, it means it denotes an Indian, Algerian, Tunisian, or those from Middle East for example… those whose skin colour is not terribly dark. Funny, because….

I interrupt her to ask who uses this term. She continues: “Polish people in England… The first thing we learnt here [in England] was ‘oh, here comes ciapaty’” and her partner adds:

I don’t know, I thought that ciapaty is a strange colour, such undefined […] and the best thing is that ciabatta⁶ is simple a kind of their bread, so perhaps that is where it comes from?

They go on pondering the term and cannot decide whether the term denotes Indian and other people who eat chapatis frequently, or whether the term denotes, as they say, their “undefined” skin colour. Notwithstanding the original intention, this neologism is used by Poles in England to represent mostly South Asians but also people from the Middle East (cf. Gawlewicz 2015). It is hardly surprising that Poles seek a new term—a category—to describe people they encounter and people who do not fit into their imaginary of racialised distinction. Public discourse on race and racism in Poland focuses only on Black people; respectively it creates a white-black binary. In the British context, this understanding of racial categories is altered and extended by a ‘third race’ defined by their unspecified skin colour (Nowicka 2018). The lexical analysis of the interviews showed that the various labels depicting ‘Indian’ are clustered with the labels ‘white’ and ‘black’, and thus belong to a racial classification scheme. The narrations around people of a ‘third race’ often include negative stereotypes that function within a racialised hierarchy, as this quotation⁷ demonstrates:

[…] they [Indians] are like this, they are different, for me worse, but at least not aggressive. They have less aggression in them than blacks who come from Africa. Because I don’t speak of blacks born in Europe, they have a different attitude, but those immigrants from Africa, Somalia, their life is a computer game, you can kill, and the person will be reborn. This is their attitude. The lives of others do not mean anything to them. (Nowicka 2018: 831)

And another participant adds:

[…] it does not disturb me if a person is black if this black person is nice, educated, intelligent, smart and resourceful, well organised and so on, and not every black is lazy, inane, arrogant, impolite, but I simply have exactly such black colleagues; India, this is such a nation that is convinced they have to get everything for free and others ought to work for them. The English got into this trouble, forget the blacks but those Indians, Muslim, they all flood this place […]. (Nowicka 2018: 831)

As Shimada (2007: 114) notices, “the gaze at the other culture is fixed long before the encounter”. So are the Polish migrants’ narrations about people they encounter in England’s cities moulded by understandings of racialised others that function in Poland (Strani and Szczepaniak-Kozak 2018; Adamczak-Krysztofowicz and Szczepaniak-Kozak 2017). Yet studying Polish migrants in England over time allows us also to observe how their racial vocabularies and attitudes transform within the British social context. When I interviewed Bianka in 2011, she told me she must remember not to use the word ciapaty but Indian instead; she signalled her awareness that the term is pejorative and improper although this does not make her change her mind about these people and assigning them, irrespectively of their religious, gendered, and class heterogeneity, to a group of ciapaty. In the later interviews (2014–2018), the term ciapaty appears seldom but racialist stereotypes about Indian, Pakistani, Bangladeshi, and other groups are frequent in the narrations. They are particularly frequent in Birmingham, and less so in London. A similar ‘third race’ category is absent from narrations by Polish migrants in Munich and Berlin in Germany (Nowicka and Krzyżowski 2017) which points to the meaning of local racial hierarchies and how they shape migrants’ acquisition of cultural codes. Similarly, Poles in the UK use ‘Asian’ to describe people from South-East Asia even though in Poland ‘Asian’ refers to Japanese, Chinese and Koreans, primarily. The same, ‘Polish-like’ understanding of ‘Asians’ occurs among Polish migrants in Germany.

How the interview participants see others is thus deeply embedded it the racialised cultures in which the Polish migrants function. Coming from a culture where race is perceived as binary, they struggle to establish how to classify those whose skin pigmentation is, in their eyes, neither white nor black. Lacking the knowledge on how to recognise people of Indian, Bangladeshi or other origins, yet sensing the need for racial classification in Britain, they invent ‘a third race’ as a category based on their visual experience. As this category is linked to mostly negative stereotypes, it becomes also an instrument of power and helps Poles establish social proximity to the white British majority (Fox and Mogilnicka 2019).

The narrations including the neologism ciapaty are also a useful example of racial othering within a culture of (visual) essentialism and racialised binaries. This is not to say that Indian, Pakistani, or Kurdish people are ‘ambiguous’ because of their skin pigmentation but that Poles perceiving their bodies feel ‘lost’ as they do not neatly fit their idea of two essentially different, thus separate, ‘races’. The presence of bodies that do not fit into a pre-imagined category shakes the general psychological essentialism (Medin and Ortony 1989).

In the western hemisphere, more and more people apparently are perceived as not easily classifiable into a white-black racial scheme.⁸ Multiracial people appear ambiguous only when racial classification is arbitrary, dichotomous and binary, an ‘either-or’ question. The presence of so-called multiracial or mixed-race people do not fit official monoracial classifications such as the national Censuses (Aspinall and Song 2014) or the social scientific understanding of visible minorities constructed through their difference from the white majority (Song 2020). This has also resulted in a new branch of psychological experimental research which tries to understand how people make categorisations of multiracial individuals, depending on their own racial identity, degree of exposure to multiracial people, and the social context of the encounters (Chen and Hamilton 2012; Freeman et al. 2013, 2016; Lamer et al. 2018). Generally, those outside a particular group might disregard the internal heterogeneity of a group: for example, people who do not identify as multiracial may assign people to a general category ‘black’ while its members may have a differentiated identity as mixed-race people (Feliciano 2015). Monoracial, in particular white people, are more likely to miscategorise those who are multiracial (Herman 2010). Research shows that people who are ‘racially ambiguous’ experience different forms of discrimination; they are more likely to be stigmatised (Grier et al. 2014) in institutional contexts such as within the criminal justice system (MacLin and Malpass 2001). They may be rejected by monoracial white and Black groups simultaneously (Campion 2019; Khanna 2010; Ali 2003; Song 2017; Chaney et al. 2020).

Others are therefore treated as surfaces reflecting one’s own classification schemes and stereotypes. This process is not accidental but deeply rooted in one’s cultural repertoire which governs which aspects of someone’s appearance is foregrounded: it could be gender, or someone’s light or dark skin pigmentation, body shape and bodily movements, or other features (Nowicka 2018). For example, to white Polish migrants, whiteness of others is a norm, something obvious and therefore backgrounded, whilst the stereotyping of Muslim men as patriarchs, machos, or ‘pashas’ makes Polish migrants dismiss the possibility of their queerness. Clearly, such categorisations are flawed and errant and often misunderstand the self-identifications of others. They are embedded in social structures of power and reflect them. Disavowing physical attractiveness when describing lower class or Black women, Polish white female migrants position themselves in a racialised hierarchy in today’s Britain and signal their aspiration to belong to the white middle class. Importantly for our book, these examples demonstrate how such positionalities rely on seeing and being seen. And, even more importantly, the study of Polish migrants tells us also that not only how we see others (stereotypes) but also, what we see in others (perception) can change when confronted with unfamiliar circumstances.

Social sciences make it clear that we live in a world in which our chances of being and becoming are largely structured by how other people see us. For example, it is well documented that women wearing headscarves experience discrimination in hiring processes (Leckcivilize and Straub 2018), obese applicants receive fewer call-backs for job interviews (Rooth 2010), and people perceived as more attractive and beautiful tend to earn more (Parrett 2015; Doorley and Sierminska 2015). Science also relies on appearance: migration scholarship, for example, frequently addresses people of colour as ‘visibly non-white’ minorities (Song 2020), using visible appearance as a marker of racial identity. Part of medical and epidemiological research routinely employs classifications referring to appearance and thereby perpetuates and stabilises racism (Hunt and Megyesi 2008; Beaudevin and Schramm 2019).

As Kress and van Leeuwen (1996: 168) write, “seeing has, in our culture, become synonymous with understanding […]. The world ‘as we see it’ […] has become the measure for what is ‘real’ and ‘true’”. Thus, we ignore that the world we feel to be real becomes so only through “the continual recurrence of identical, familiar, related things in their logicized character” (Nietzsche 1968: §569). Yet the ways we are seen do not do justice to us, in many ways. People suffer when some of their features which, to them, are vital to their identity are overlooked by others; or when they are reduced to only a few characteristics and the complexity of their personality or positionality remains unrecognised. ‘Looking’ in all its different forms and modulations can be a powerful tool with which to build or sustain privilege and exclude particular individuals, or indeed, entire groups (Brighenti 2010). Being an object of another’s gaze defines the lived experience of people “locked” in such reification (Fanon 1986: 89ff). On the other hand, some groups may use the visibility of some aspects of their identity or positionality tactically, for example in the political struggle for recognition (Spivak 1988; Narayan 1997). When and if strategic (visual) essentialism is legitimate is a subject of controversial debates (Pande 2017; Stone 2004; Hoyt et al. 2019).

How we are seen is not exactly who we are or want to be. Divergences between self-perception and other’s classifications are well documented in the literature on race relations. They may emerge due to different cultural conceptions of a category; for example, ‘black’ in the US-American, British, or Brazilian context may be understood differently and associated with different features. Also, general and scientific beliefs of race may diverge (Suyemoto et al. 2020). Similarly, the Western popular concepts of gender as a binary, commonly associated with physical characteristics of bodies, is at odds with a scientific, more nuanced, understanding of sex as a spectrum (Ainsworth 2015) as well as with non-Western knowledges (Johnston 2005). There is also historical evidence that boundaries between categories such as white and Black vary across time and space. In his influential book How the Irish Became White, the historian Noel Ignatiev (1995) describes how Catholics from Ireland who immigrated to the US in the nineteenth century where subjected to discrimination, not dissimilar from that experienced by the Black population there; initially, they also identified with African Americans, but with the changing demographic and political situation in the US, they slowly gained the privileges, rights, and duties of white Americans. Similarly, Eastern European migrants in the west of Europe struggle to gain recognition while their whiteness is under scrutiny (Tereshchenko et al. 2019; Paraschivescu 2020; Fox et al. 2015). These examples demonstrate that whiteness or Blackness encapsulates more than just skin colour; it reflects a complex system of privilege and disadvantage which emerges and develops in particular historical and geographical contexts. But these illustrations also show that social visibility as a racialised other is embedded in particular contexts and political, social, and technological arrangements (Brighenti 2010; Song 2020). For example, whether someone is considered Black or poor can depend on where the encounter takes place. Recently, Ludwig and Kraus (2019) demonstrated that some features of a neighbourhood, such as degree of ethnic diversity or wealth, may impact people’s judgements on other’s and their lives. This means that people are read through the contexts in which they appear. Visibility is thus an outcome of a complex social process which locates people in particular relational positions (Anthias 2008).

In this sense, social relations are inscribed on bodies. Despite the fact that historians of race relations (as discussed above) could convincingly demonstrate the historical and geographical variation of social categories, this knowledge is somehow contradictory to the “western modern imaginary” (Taylor 2007) which makes us believe that the body is the expression of the subject’s “psychic interior” (Grosz 1993: 198). It means that ‘race’ or ‘gender’ is seen as biological (natural) difference and not merely a projection of social relations on bodies, or embodied social relations. Our eyes and our brain seem just to ‘reflect’ the ‘true’ nature of the bodies they encounter. Seeing is believed to be an act of ‘absorbing’ existing human difference. Oyěwùmí (1997) reminds us that privileging biological explanations of differences of race, gender, or class is historically new and geographically limited to European cultures. Other cultures do not privilege the visual in the same way. The reliance on seeing as the direct path to knowing is thus a typically western mode of understanding. As infants in western modern culture are asymmetrically exposed to different types of faces, including female, male and other genders, and more to their own-race than other-race (Anzures et al. 2010; Quinn et al. 2018), children by the age of three to five develop racial biases (Gelman 2003; Hirschfeld 1996). Accordingly, this racialised way of seeing is perpetuated intergenerationally.

Race and gender seem to be known by observable physical cues that inhere in bodies and they seem to be experienced unmediated by culture, that is, they gain their salience from their self-evidently striking nature (Obasogie 2010: 586). They are social categories that have a strong visual character and tend to be essentialised (Bastian and Haslam 2007). Yet, the visuality of race and gender is not limited to immediate visual perception of the observer. The imaginary of race and gender as visually perceptible and discrete categories is also linguistically anchored (Orians 2018). Visual metaphors are common in our languages: we look at a problem, we see the point, adopt a viewpoint, focus on an issue, and see things in a perspective (Kress and van Leeuwen 1996: 168). Such metaphors link semantic representations and cognitive structures (Wilson and Foglia 2017) and affect how we act (Lakoff 2014). By visual language, even blind people learn that race is a matter of visible features, and they learn how to align their sensory experience of others to the visual racial scheme (Obasogie 2014).

Being in the world and experiencing it both mentally and corporeally means, that we think, feel, and behave in complex ways which can hardly be articulated in words. The language which we have at our disposal to make sense of this experience, as humans and as social scientists, relies heavily on discrete social concepts such as ‘woman’, ‘short’, ‘person of colour’, ‘student’, ‘mother’, ‘transgender’.

The terms I could use to describe myself are of a different nature. Some relate to my social role—mother, researcher. And some refer to my appearance—for example, short. I identify as a woman, and most people identify me as a cis-woman after assessing the shape of my body, the timbre of my voice, and the way in which I dress which they associate with femininity. I am below average height for women in Germany, but of average height in Japan. Whether I am always white, is another complex story (Fox et al. 2012; Böröcz and Sarkar 2017; Bonnett 1998). As a person born in Eastern Europe, my whiteness in present-day Western Europe may be contested (Lapiņa and Vertelytė 2020). Thus, being white or short is relative to the context in which I live.

I can say that my experience is that of a rather short, white woman who is a mother and researcher, but this does not capture fully how I experience the world. Moreover, some of these categories are highly political, and if I describe myself as white, this is not just an indication that my skin pigmentation is rather light but could be understood as an expression of a political identity. As whiteness for so long has been the unacknowledged norm because of white supremacy, if I do not articulate that I am white, my counterpart might assume it. Nevertheless, by acknowledging the intersection of these categories, I can try to express that being white inflects my being a woman, to claim simply that I am a woman, denies the particular experience of white womanhood and how it affects ‘being a woman’. Or, in other words, there are many parts of my identity that are reliant on others: my experience of ‘being a woman’ is influenced by the fact that I am a mother; my experience of motherhood is influenced by the fact that I am white and so on. These parts of me are inextricably linked and cannot simply be separated and observed side-by-side. I need to establish a conceptual relationship (Avargues-Weber and Giurfa 2015: 3) between these categories to describe myself in an intersectional way.

Yet even the intersections of categories do not fully capture the complexity of experience, for each category tends to be too large or too narrow. One can be a woman in a myriad of ways, and white or black in a myriad of ways as well. It is the intra-categorical heterogeneity which challenges the oversimplified description of human experience. While it seems obvious that being a woman, or being white, or being a mother, or heterosexual is multidimensional, assigning something or someone to a category means to suppress the difference within this category and exacerbate its difference from other ‘categories’ in this topological ‘regional’ thinking (Jraissati 2019; Mol and Law 1994). Therefore, categories are always contested, and their boundaries and meanings are re-negotiated. But some categories seem more persistent, as they rely on visual evidence—I would probably be perceived as a white woman anywhere but hardly anyone recognises me as a researcher based on my appearance alone.

Intersectionality research gives us a good sense of how the experience of discrimination is irreducible to the workings of one category. As Patricia Hill Collins (2019) argues, intersectionality has always put relationality in focus. The relationality of categories could be understood in a cumulative way, in which one category is added to another, altering the meaning of both (Collins 2019: 229). Relational thinking could also be defined in terms of articulation, juncture, or connection from which two entities create (for a time) a new entity, quality, or pattern. Racism, sexism, capitalism, and homophobia are articulated differently in and across varying social contexts but under certain circumstances they cohere (Collins 2019: 233; Hall 2017). Co-formation as a third type of relational intersectional thinking “seemingly dissolves the categories themselves” (Collins 2019: 241), thereby Collins stresses that this kind of dissolvement is an intellectual and theoretical exercise enabling the development of a holistic analytical framework, for in reality, categories do not dissolve. The aim of intersectional analysis through co-formation is to adequately acknowledge the complexity and irreducibility of the experience in the world, thereby intersectionality research must resist the western modern mode of producing scientific knowledge through dissecting and disaggregating phenomena it studies (Collins 2019: 244).

Collins warns that relationality understood as addition bears the danger of adding categories of experience to each other instead of incorporating a category of analysis into the study (Collins 2019: 227–228). The first move corresponds with what she calls the western logic of segregation. The second means to address how one category is changed when it is added to another. For example, an analysis that starts with gender and ‘adds’ race and moves to include sexuality, ability, and so on engages with social experience differently than if ‘race’ or ‘ability’ are a starting point. The remainders of this chapter involve insights from cognitive science to discuss the possibilities of intersectional categorisation that could account for relationality and fluidity of human experience and identity.

This chapter focuses on race and gender as categories of experience. The decision to exclude class, despite the fact that class is—next to race and gender—a central concept in intersectionality research, in particular in the Marxist tradition of conceptualising race relations, is motivated by several considerations, some conceptual, and some pragmatic. Class designates both one’s social and economic position, and an aesthetic and affective disposition (Bourdieu 1984, 1977). An understanding of class includes hierarchical categories of wealth, education, occupational profile, and lifestyle. At the same time, class is an analytical or a descriptive concept used to analyse capitalism and its workings or to describe identities. Visual indicators of wealth, occupation, and education (signalled by lifestyle, clothing, etc.) serve to signal one’s social proximity or distance to others. Next to institutionalised forms, this form of distinction is used to exclude members of other classes and secure one’s own status and privileged access to resources. Class distinction intersects with gender and race; yet it is also widely believed that people can be mobile across the class hierarchy (Varnum 2013). Indeed, the concept of a middle class is based on the potential class mobility—aspiration of its members to move upwards, and their fears of downward mobility characterise the middle class. In particular, intergenerational class mobility makes the category of class appear more ‘social’ and less ‘biological’ or ‘genetic’ (Kraus and Tan 2015).

While actual class positions continue to structure social relations, and people’s visual perceptions of others (Harrits and Pedersen 2018), Collins (2019: 230) argues that class is an underlying master concept (and not a category) for any intersectional research. Adding class to an analysis of gendered and racialised processes and interactions is in fact unnecessary, for class analysis—respectively the analysis of capitalism—must instead prefigure intersectional analyses.

Current research in cognitive science suggests that categorisation is a basic form of interaction between humans and their environment. Research in the field of neuropsychology, cognitive science, and psychology shows that the world we see is not equivalent to the physical world but ‘biased’ (in a predisposed way) according to our individual sensorial abilities, such as contrast sensitivity or colour and motion perception, and sensory experiences. The most obvious examples include various forms of colour blindness, or different forms of impairment of perception of depth and three-dimensional structures. But our perception is ‘biased’ also in other ways which are more relevant in the context of our book.

To clarify, it is useful to distinguish between perception and cognition. Broadly speaking, perception is what puts us in contact with our present surrounding by ‘analysing’ sensory experiences, while cognition is what enables us to form beliefs, make decisions, to act. There is a discussion in psychology on the extent to which perception is penetrated by our cognition (Firestone and Scholl 2016; Halford and Hine 2016; Montemayor and Haladjian 2017; Vetter and Newen 2014). The research on perception in space, for example in complex multidimensional settings such as a street crossing, shows that cognitive mechanisms may guide our perception; we are attentive to some but ignore other visual stimuli, which is a cognitive effort (Niv et al. 2015). Research on infants proves, on the other hand, that perception of colours as distinct from each other (categorical perception) emerges before infants begin to develop concepts of colour (Franklin 2016), suggesting that perception is indeed separate from cognition. There is further evidence to suggest that the time between seeing and categorising colour or shape is extremely short; it allows for the conclusion that a distinction between perceiving and cognitively categorising an object is not straightforward (Cichy et al. 2014). Ultimately, scholars have come to agree that the demarcation between perception and cognition might be blurry.

The interest in perception is by no means an exclusive domain of cognitive science and psychology. The idea that our perception is not always veridical—a true representation of physical reality—has also been taken up in philosophy (Crane and French 2017). The question philosophers ask is how do we make judgements and form beliefs about the external world? Is there a reality that is completely independent of our thinking? Do we perceive or experience reality directly, or indirectly (through ‘the veil’), and thus cannot know what ‘true reality’ is? Is the nature of perception different from the nature of thought (cognition)? The answers to these questions have epistemological consequences in at least two ways: first, if we, humans, all perceive objects similarly (providing no neurological differences) and independently of our thinking, then we should indeed easily agree on what we see, for we all see the same; secondly, if perception is independent of our beliefs, our perceptual experiences can be seen as a cause and as justification of these beliefs (Smithies 2016). Deroy (2013) gives a simple example: if an apple we see is red and our perception of colour is independent of our thinking, then if one person sees this apple as yellow, then the difference in perception must be due to the person’s neurological condition. The rest of us will easily agree that the apple is red. Moreover, our belief that most apples are red is determined by our perceptions of apples as red. Yet various experiments confirm that having learnt that apples are red, and bananas yellow, we are more likely to perceive shapes resembling apples as being red and banana-like shapes as being yellow. Our idea of red apples thus influences our visual perception.

Cognition is largely about categorisation (Harnad 2017). Categorisation is a mental operation by which the brain classifies objects and events (Cohen and Lefebvre 2017: 2). Any category includes kinds of objects; a category has both an extension (the set of things that are members of that category) and intension (the feature that makes things members of the category rather than another category). From an ontological point of view, all things are members of an infinite number of different categories, and each of their properties/features and their combination is a potential basis (affordance) for assigning the thing to still more categories—(Harnad 2017: 31)—I call this “ontological” intersectionality which is distinct from heterogeneity within one single category.

Recognition is a form of interaction with things. Unlike sensory interaction (through seeing, hearing, smelling, tasting, touching) or naming and describing things—recognising things means to see something as a kind of thing that we have never seen before—that is, as a member of a category. To recognise, we need to abstract, that is to single out some subset of sensory (e.g., visual) input, and ignore the rest (Harnad 2017: 34). To abstract means to detect recurrences and to ignore the uniqueness of things and the particularity of their contexts. Through abstraction, we recognise sameness or similarity, and identify kinds of objects, states, events, or individuals.

We abstract things that have properties, meaning the features of sensory input that we can detect selectively (Harnad 2017: 36). The interactional approach, following Gibson (1979) conceives of such properties as not simply features of a thing but rather ‘affordances’, that is “the ways in which things come into the immediate presence of perceivers, not as objects-in-themselves, closed in and contained, but in their potential for the continuation of a form of life” (Ingold 2018: 39). With her proposal for the metaphysics of categories, the philosopher Ásta (2018) argues in a similar manner. She assumes that objects have various physical properties (physical facts); according to Ásta, these are of two kinds: conferred and unconferred. Some (base) properties of an entity may remain unconferred by our judgements. But if we make a judgement on the property, we confer it. This means the entity gains a particular meaning in a specific socio-cultural context in how it is conferred by others. Some properties are obviously recognisable as conferred, for example a person’s “popularity” is clearly ‘not natural’ but dependent on a relation someone or something has to other people (Ásta 2018: 8) but other properties seem more ‘natural’ and unconferred, while they, in fact, are not. Gender or race could be such seemingly unconferred, but in fact conferred, properties.

The cognitive process includes two different operations, discrimination and categorisation (Harnad 2017: 38). Ontological intersectionality can be understood as a spectrum of (visual) impulses. We ‘discriminate’ between grades through direct comparison, so that we can decide on similarity and dissimilarity of things when we see them (just-noticeable-difference). ‘Categorising’ means to decide on (dis)similarity of two things in isolation, it is to identify something as bigger or smaller, lighter or darker, based on our memory of the other thing. It means, we need to have a concept of something to categorise what we see (Jraissati 2019: 423). The human capacity to categorise stands in relation to memory capacity and is thus limited (Cowan 2015); in consequence, we categorise things which are very different from one another better (more correctly). This recognition is of importance when we consider ‘biased perception’, for example, ‘racial perception’.

Categorising is a learning process. It includes mapping sensations (received by our senses) onto conceptual spaces (Jraissati 2019). As we learn to categorise things, we also learn to exclude the alternatives, that is what does not belong to a particular conceptual space. If ‘affordances’ of things are obvious, such as the difference between a floor and a wall, learning how to categorise two surfaces might be easy. If ‘affordances’ are not immediately clear, learning to categorise requires more effort (Harnad 2017: 43). Naming things (language) helps us to categorise (Folstein et al. 2015; Jraissati 2019). This is the role of culture in which we live and which is based on language: it helps us to deal with ambivalences and uncertainty with respect to categorisation. Without culture, categorisation would be based exclusively on sensory and trial-and-error methods, which involve consciousness. Categorising correctly has one consequence, miscategorising another; of this, we must become conscious. Sensory based learning is trial and error, reward and punishment, through which we get better at categorising correctly. We can think of an analogue to a tennis player who learns consciously how to hit a ball the way it passes the net, and how many attempts the player needs to do it (almost always) correctly, before they become unaware of each movement they make. Such learning is less efficient—in the sense that it takes more time to change the practice—than learning with the help of culture which ‘tells us’ who is who and how to assign a person to a category making ‘fewer mistakes’. Experiments increasing consciousness might, therefore, not only be ineffective in reducing racial prejudice but even increase it, as long as they happen within a racialised culture (Strick et al. 2015). At the same time, as language impacts visual perception (Jraissati 2019), it is possible that in cultures with language including, for example, a separate word for multiracial individuals, these people are less likely to be miscategorised.

A small number of categories could not function without grounding them in sensory experience; other categories function through association, or, in other words, they acquire their meaning through reference to other categories. ‘Race’ is such a category which is independent of us seeing differences in humans. The variation within ‘racial’ groups is greater than it is between them (Witherspoon et al. 2007) as common phenotypic markers exist on a continuum. In this sense, there is no physical or biological difference between ‘races’. When we categorise people as belonging to a racial group, we do it with reference/association to several other categories which are grounded in sensory experience such as skin pigmentation, hair texture, eye shape. The closer an ‘associative category’ is to something that we can directly perceive with the senses, the more concrete, real it seems to us (Vincent-Lamarre et al. 2016: 650). This feeling of being ‘true’ because something is grounded in the sensory experience can be strengthened through emotions (Kousta et al. 2011).

Recognising human difference depends on various socio-cultural factors, which is well documented for the so called ‘other-race’ or ‘cross-race’ effect.⁹ It is a socio-cognitive phenomenon which means that people more easily recognise faces of those who belong to their own ‘racial group’ (Young et al. 2012). It is not quite clear why this is; limited experience could be responsible for this effect (Rhodes et al. 2010). Infants recognise human faces differently depending on their gender (Ramsey et al. 2005) and ‘racial characteristics’, and this seems to be due to children’s asymmetric exposure to people visually different from their parents (Slone et al. 2000; Lee et al. 2017). It is also known that if adults are accustomed to seeing people whose racial identities differ from their own, they are more likely to recognise the face of an ‘other-race’ person quicker and better (Levin 2000; Walker and Hewstone 2006). If someone is socialised with people whose identity is similar to their own, the ‘cross-race’ effect could be supported by the frequency or repetition of information (Kahneman and Tversky 1972). Thereby both gender and ‘race’ seem to matter at the same time. Masculine and feminine faces, for example, attract our attention differently quickly and for a different period of time; also, women tend to attend to ‘own-race’ women longer (Lovén et al. 2012). There is some evidence that those who harbour racial prejudices are slower when it comes to recognising the faces of people whose ‘race’ is different from their own (Costandi 2012); reducing prejudice could minimise the ‘cross-race’ effect. Also, we tend to see people who belong to a group to which we also count ourselves as more heterogeneous than members of an outgroup (Meissner and Brigham 2001).

As previously mentioned, categorisation relies on suppressing difference within and exaggerating the difference between people to enable the construction of groups. Following Searle’s (1996) account of the construction of social reality, we can say that in the process of categorisation, somebody becomes what they were not before: through culture, a body becomes a racialised or gendered, young or old, abled or disabled body; skin becomes white or black; the shape of nose or eyes becomes a marker of belonging to a group.

For a long time, psychology maintained that categorisation necessarily results in stereotyping. Stereotyping is linking additional information to a category. As both processes—categorisation and stereotyping—are largely implicit and include oversimplification, it was difficult to prove their distinctive nature. But newer research suggests that these two processes might be independent though interrelated (Ito and Tomelleri 2017). It seems that categorisation results in stereotyping only upon the existence of other enabling circumstances. A popular account says that stereotypes deviate from the ‘true’ or ‘real’ features of the person or a group. Stereotyping is ‘easier’ if it focuses on the most different or ample type from a group, one that represents it best. It thus is also easier when difference is expressed as a binary contrast, for example black and white. Stereotypes ‘anchor’ better on such most highlighted types.

Stereotyping could also be understood as the idealisation of a category: for example, the category ‘grandmother’ includes people who are female and mothers of a parent; a typical grandmother, though, includes additional information (context-specific), for example: white hair, rocking chair, bakes cookies. If a person’s features do not correspond with this stereotype, we might have problems categorising her as ‘grandmother’ or might miscategorise her. Similarity to the stereotype in this sense provides probabilistic information about the person’s membership in the category (Hampton 1998).

Stereotyping could be considered the mechanism which translates categorisation into social hierarchy, particularly if a stereotype is negatively loaded, or when stereotypes lead to largely invalid/assumptive expectations of a person’s behaviour. For example, associating women with acts of care and love might result in a gendered expectation of women, casting them as stay-at-home mothers, or limiting their occupational horizons as those engaged largely with care-work. Instead of thinking of stereotypes as a biased idea of people or groups, newest cognitive research more forcefully acknowledges the role of culture, observing that, for example, stereotypes have different cultural value (Hinton 2017). In this sense, a ‘biased view’ means a view that diverges from the hegemonic position (social norm) rather than from a ‘true’ condition, quality, or characteristics of a person (or a group, an object, a situation). Stereotypes thus reflect a history of discrimination and domination of some groups over others; racial stereotypes, for example, are rooted in histories of enslavement and geopolitics of exclusion, and gendered stereotypes reflect a long history of male domination (Hernando 2017).

Researchers have been interested in how stereotyping, whether ascribed to people or objects, positive or negative, is an effective tool in rapid decision making (Bordalo et al. 2016). It has been argued that implicit associations have an evolutionary basis and bring about a survival benefit (Fox 1992). This does not mean, of course, that the current content of stereotypes (e.g., on gender or race) is good or necessary, but that stereotyping can be an effective way of dealing with the complexity of the world, a feature that Devine and Sherman (1992) call a ‘cognitive economy’. The mechanism used in stereotyping is described by Tversky and Kahneman (Kahneman and Tversky 1972, 1973; Tversky and Kahneman 1974) as representativeness heuristic. Heuristic is a practical method, a ‘mental shortcut’ used in making rapid decisions and solving problems. The representativeness heuristic relies on assessing similarity of objects and organising them around the category, according to the assumption that like goes with like, and thus causes and effects should also resemble each other. Stereotypes could be thus defined as a human brain strategy for coping with limited information processing capacities in a complex social world.

The question is how to change implicit associations that lead to social injustice, discrimination, and exclusion if the mechanism apart from stereotyping is so central to our being in the world. One way to do it is to try to replace one association in semantic memory with another one. For example, a negative association could be replaced by a positive or neutral one. As this is done through a conscious training, it is less efficient (or cognitively more costly) than unconscious associating, the effects of such trainings are not durable (Lai et al. 2016). Neither explicit reflection on one’s own racial privilege (white privilege checking) nor learning of novel associations and erasing of the old ones, proved so far to be an effective way of changing stereotypes (Kalev et al. 2006; Boatright-Horowitz et al. 2012). Yet the content of stereotypes changes over time, albeit slowly, and therefore some researchers suggest that to change stereotypes requires a change of the culture in which a stereotype is embedded.

A more viable possibility is suggested by research on the ‘predictive brain’. Clark (2013, 2014) suggests that attention is functional, and that human perception is predicting. The dynamic brain predicts the experience and ‘compares’ the prediction with reality; if the expectation was incorrect (does not match the reality well), the brain will ‘correct’ it and predict better next time. To minimise error, the brain uses probability calculations; these are culture specific and expressed as stereotypical knowledge (Hinton 2017). Following this trait, we could speculate that people frequently confronted with unpredicted, surprising occurrences could adapt their stereotypes. Each time their prediction turns out wrong, their brain ‘recalculates’ and revises the prior prediction to get better next time. This mechanism could be used by artistic research, or arts more generally, to increase the capacity for intersectional seeing.

Intersectionality poses a challenge to prevalent theories of stereotyping. Intersectionality research seeks to contest essentialised and fixed group categories as a mode of understanding our experience in the world; stereotyping is the process of establishing essentialised, exaggerated conceptions of individuals and/or groups. Therefore, the two concepts might be incongruent (Cassese 2019; Remedios and Sanchez 2018). The major problem of research on stereotyping is that it operates within a particular historic and cultural context which prioritises discrete social categories. One result of this is the compartmentalisation of research on prejudice and stereotyping, it is the focus on one domain or category, such as gender or race (Bigler and Liben 2006; Ghavami and Peplau 2013). Some research tries to overcome this limitation by manipulating one category (e.g., gender) while keeping the other constant (e.g., race), and in this way testing intersections of these two categories. Only recently, psychology addressed how stereotyping is complicated by the fact that people belong to multiple social groups simultaneously (Cassese 2019; Petsko and Bodenhausen 2020). This research acknowledges earlier studies which demonstrated that stereotypes of black men in leadership roles may change from negative to positive depending if these men are heterosexual or not (Wilson et al. 2017), or that black individuals are judged more negatively when they are young as opposed to old, which suggested that single-category-based evaluations are contingent on another classification (here, race and age) (Kang and Chasteen 2009).

Yet the problem remains that this research nevertheless relies on a binary gender scheme (male-female) as well as a fixed category of race (usually dual, as white-Black, or triad—white-Black-Asian) irrespective of the context of research and fluid nature of racial distinctions, or the gradual character of skin pigmentation. Moreover, the experiments usually focus on faces, and their results therefore are hardly applicable to real-life situations when people ‘reinterpret’ the information vis-à-vis information on body shape, clothing, and so on (Remedios and Sanchez 2018). Also, emotions may impact categorisation (Brooks et al. 2018), and thus also intersectional stereotyping. Further, studies frequently manipulate stimuli (such as pictures of different faces) with the help of software which itself might be racially biased (cf. Chap. 4).

Notwithstanding these major shortages of the psychological research on intersectional stereotyping, the debate offers some interesting observations. So far, several possible mechanisms of intersectional stereotyping have been suggested. Firstly, it seems that targeting subjects by stereotypes may proceed by all detectable social identities of the target at once; for example, a person could be detected as ‘white young woman’ and not as ‘white’, ‘young’ and ‘woman’ and a stereotype would be integrated (Ghavami and Peplau 2013). It presumes though, that a person possesses a premade integrity. Secondly, stereotyping might focus on a certain social identity (possibly because of the perceiver’s own proximity or distance to a target of stereotype and this particular identity). One of many categories would be thus a focal category. For example, a woman would always foreground another woman’s identity as a woman, ignoring or deprioritising her other identities. Accordingly, she would stereotype the other woman with gendered/female content rather than racial content. Thirdly, social context may inform which intersecting identities of the target are emphasised by a perceiver, or even recognised. It could be the intersection of gender and race in one, and of race and age, in another social context (Petsko and Bodenhausen 2020).

Similarly, Hall et al. (2019) propose in their model of stereotyping that categories may be implicitly connected (associated): either on the basis of their phenotypic similarity or stereotypical overlap. In the model of phenotypical similarity, a perceiver can link two or more categories to shared appearance attributes. Race and class, race and gender, and ethnicity and religion seem to be such associated categories (Hall et al. 2019). Researchers could demonstrate that Sikhs are ‘mistaken’ for Muslims because of their phenotypical similarity to people who are Muslims. As such ‘mistakes’ became more common in the aftermath of 9/11 (Jhutti-Johal and Singh 2020), a geographical (Pauker et al. 2018) and historical context seems to play a role here as well. In the stereotypical overlap, two or more categories ‘share’ a stereotype, which leads for example to misperceptions of black and Asian faces (Stolier and Freeman 2016; Ahluwalia and Pellettiere 2010). Johnson et al. (2012) suggest that a racial phenotype, for example ‘Asian’, shares features with a female phenotype, and a ‘Black’ phenotype with a male phenotype. In turn, Asian women would be recognised faster as women than Asian men as men, and so on. Yet, so far, this research could not establish the direction of causality of this process, that is, whether ‘facial characteristics’ or stereotyped attributes shape intersectional perception (Kim et al. 2015).¹⁰

If some categories or their intersections are foregrounded in a certain social context, as suggested by Petsko and Bodenhausen (2020), the question is, what causes perceivers to attend to some identities and not to others? Petsko and Bodenhausen (2020) suggest four possible reasons for foregrounding: category accessibility, perceiver goals, category fit, and category distinctiveness. A category which is more easily retrieved from memory (is more accessible) may be preferred (foregrounded) despite the intersectionality of the target. For example, perceivers with high levels of racial prejudice are more likely to use race and not gender to stereotype a person. Or perceivers may accentuate this category which is better aligned with their goals than another category. Someone ‘focused’ on women for some reason would rather ‘see’ women among people of different skin pigmentation, than men. A third possibility is that a category itself may fit the social context better than another one, in normative or comparative terms. Finally, more distinctive categories could be more attention-grabbing: among all white men, a white woman would be more likely to be perceived as a woman (gender differentiates her from the group), in comparison to a situation when a non-white woman is surrounded by other white women (here her gender is not a distinctive category). Finally, it seems that people tend to categorise others slower if requested to ‘order’ the targets to intersectional (e.g., Asian woman) rather than to a single (e.g., Asian, or woman) category (Remedios and Snyder 2018). All of the models of intersectional stereotyping considered so far suggest that attention, salience, and exposure matter for how people are perceived.

Attention is necessarily selective. The process of focusing attention on something or someone implies exclusion; inattentiveness is thus an effect of being attentive. It means also that something or someone is visible in effect to the extent to which the viewer attends to, or, in other words, that visibility is a function of attention (Zerubavel 2015: 2). We are not necessarily conscious about what we attend to and what we ignore, but attention can be socially trained. Attention is thus a socio-mental act. Zerubavel (2015) claims that attention and inattention are conventionally delineated. Thus, people are members of attentional cultures and subcultures: I might be ‘trained’ in my socialisation as a girl and woman to pay attention to children as this is demanded from women in my culture, and I might be ‘trained’ to pay attention to signs of sexism as a scholar within a university subculture.

What is worth, or what necessitates, one’s attention is normative, for it is subject to a social norm, and it could be a moral imperative; as a mother I often experience moral judgements on whether I sufficiently pay attention to my child climbing a boulder wall at the playground, at least when the child is young. Me paying attention to my teenage son at a boulder wall would be less accepted or even morally rejected as exaggerated care. My gazing at a bouldering man could be interpreted as sexualised interest in this person if I ignore other boulderers, and so on.

Similarly, inattention is learnt in a social context.¹¹ Ignoring others to some degree has been described as a modern urban phenomenon, for example by the sociologist Erwin Goffman. Studying the everyday life in a city, Goffman called one pattern of people’s behaviour ‘civil inattention’, a display of disinterestedness which does not signal disregard of the other person. In this sense, inattention is more than just the lack of a gaze; it is a certain kind of social relation and a competence to refuse relations without creating non-persons (Goffman 1963). Such competence is important to establish conviviality in settings which do not allow for much physical distance but require social detachment, such as a public elevator cabin, studied ethnographically by Stefan Hirschauer (2005), a crowded metro platform, a train, and so on. Inattention, in the sense of being attentive only for a short while—or the capacity to redirect attention quickly—is a principle of modern capitalism which profits from us shifting attention from one object to another, from one product to another, and seeking not just a new object, but something new, surprising, outstanding, and extraordinary—attention-catching (Schroer 2019). Producing (visual) attentiveness thus is considered a feature of modern industries.

At the same time, there are areas in which inattention is undesired, even dangerous, and sanctioned. Crary (1999) shows that the modern highly specialised industrial production requires workers to be focused and attentive to just one particular aspect of the process. Attention, and sanctioning of distraction, goes hand in hand with increasing specialisation of professions, including scientific enquiry. A modern subject—an individual—is self-disciplined, focused, attentive (Reckwitz 2004). If it fails to bring attention, it is to be blamed for its own failures. Attentiveness is thus a (western) modern technological and economic imperative. Inattention can also be pathologised: we can think of the attention deficit disorder (ADD) diagnosed frequently in children that causes a range of behavioural patterns such as difficulty attending to instruction, focusing on schoolwork, keeping up with assignments, or completing complex tasks.

Attention is thus more than a gaze or looking. First, focusing attention is multimodal; it encompasses vision as much as other senses. By attending to something or someone, this mixed modality becomes irreducible (Crary 1999: 3). Second, attention should be conceived of as a model of subjectivity that emerged in western modernity (Crary 1999; see also Schroer 2019; Wehrle 2013). It was made possible by various social and technological developments—in this sense, as Crary (1999) argues, attention is an effect of different forces and relations of power in modernity and not simply a domain of the visual. An attentive subject, and a subject of attention, in (western) modernity is a subject whose capacity to perceive is located within the body; in turn, the attentive subject is isolated from other people and separated from the environment (Crary 1999: 3). Dis-integration of an attentive subject and the surroundings opens up the space for technical and psychological manipulations of (visual) perception (through various technologies of attraction or scientific experiments in psychological research on stereotyping). Crary (1999: 25) argues that the centring of the concept of attention in (western) modernity results in a larger set of positions and consequences for thinking of related issues; for example, understanding of attraction as a competence which can be trained and manipulated results in the idea of human perception as ‘impure’: any sensation can thus be understood as compounding of memory, desire, will, anticipation and immediate experience. In this sense, perception—or specifically seeing—cannot be reduced to visuality. The difference that we ‘see’ is not the one located in the body of the other, nor is it embodied by the observer, but happens in-between the body of the other and the perceiver, an effect of relations of power. Alternative vision is constituted in modernity as another cognitive bodily state (trance, reverie, hallucination). In turn, the centrality of attention in (western) modernity disempowers the subject, stripping it of the power to act and the power to resist (Crary 1999: 3).

Through the lens of attention, the main question is how our perception of others as members of one, or more, or all or no categories is fashioned by powers external to the perceiver and to the target. It exceeds the scope of this chapter to discuss in detail all proposals for how societal values and norms are incorporated by individuals and transformed into their personal convictions. Generally, scholars in various disciplines (philosophy, sociology, anthropology, linguistics, political science, psychology, and cognitive science) use the terms ‘scheme’ and ‘frame’ referring to internal mental structures and external, ‘cultural’, patterns (Lizardo 2016b; Wood et al. 2018). One possibility to conceptualise the relation between public ‘culture’ and individual perception is to consider how social frames evoke meanings and activate schemes. The last could be understood as embodied multi-modal experience, stored in memory, nondeclarative and habituative, flexible patterns of interpretation which do not have any specific content, and can thus be transposable across variable situations (Wood et al. 2018). Importantly, such personal schemes can be altered through repeated experience, yet this process is slow (Lizardo 2016a), as suggested by the research on the ‘predictive brain’ (Hinton 2017).

My own research suggests that some situations (frames, in the terminology of Wood et al. 2018) may speed up the re-learning process, or at least disrupt personal schemes to some extent. Researching Polish migrants in Germany, we noticed that they constantly refer to other people’s ethnicity in narrating about their daily life and surroundings (Lisiak and Nowicka 2018). Ethnicity appears to be a basic operator, a practical category with which they perceive other’s difference (Brubaker et al. 2004), in particular when others’ difference is juxtaposed with their own identity as Polish. Yet their narrations also include many references to irritating situations or the feeling of a certain discomfort, or even annoyance. These narrations assigned positive or negative judgements to people’s behaviour or utterance using ethnic labels, yet the irritations were caused by a perceived mismatch or uneasiness between their own practices in a new context. Polish migrants frequently identified—but could rarely express it explicitly—a different pace of doing things in Poland and in Germany, and more generally, people’s different approach to time. This manifests in the speed at which people try to complete certain tasks, if they work over hours, how they spend their leisure time, or if they make investments in their own or their children’s future career by attending additional courses, and so on. Such approaches to time (personal schemes) respond to neoliberal imperatives of ‘good worker’, or ‘good mother’ that are embedded local contexts (frames) in Poland or Germany. By migrating between these contexts, people we interviewed experience a sense of the misalignment of their own practice to that of their surroundings. They become attentive to what otherwise remained hidden to them.

Stoltz and Taylor (2017) remind us that material context matters for attentiveness. Placing objects in an unusual way or location makes them more perceptible, exposed, and foregrounded. We could think of a luxurious car parked in a poor neighbourhood, or a black female parliament member in a parliamentary assembly room occupied by white male members. Discussing attention requires thus to consider materiality and the meaning given to the target of perception. Any purposeful enunciation or manipulation of attention relies on these two aspects: we can think of subversion, controversy, or resistance in artistic practice achieved by placing objects outside of their usual context to evoke irritation or surprise, reverse conventional attentional patters, re-direct the attention to what we habitually background, and lend the object or situation a new meaning (Zerubavel 2015: 8).

We learn from cognitive science that categories—understood as a mode of (visual) experience of the world—are an effective way of making judgements and taking decisions. With categories, humans reduce the complex sensory (visual) input and create mental representations of the world (Gärdenfors 2019). In this sense, categories are ‘useful’ to navigate the world, but also necessarily ‘flawed’. Or, they ‘hinder’ us from seeing the world ‘as it is’. Contemporary cognitive scientists no longer presume that categories are determined by physical (i.e., located ‘in the world’) or physiological (i.e., located ‘in the brain’) principles (Jraissati 2019: 422). The consensus on the lack of such nativist determinism does not mean that categories can easily be changed. Our perception is structured, and culture (including language) plays a role in how it is structured. It is a subject of intense efforts in cognitive science to understand how we decide on assigning an object to a particular category. The culturally shaped and shared concepts may ‘help us’ in this cognitive decision (Deroy 2019).

Categories create a horizon of our experience in the world, yet the experience (seeing) of the world and representation of it do not match perfectly. In this ‘gap’, there is room for change, even if we tend to rely on the supposed visual evidentiality of difference and similarity. We are attentive to some aspects and ignore others; in the culture which assigns meaning both to visual experience and to the concepts of race and gender, the principle of categorisation and the factors in social stratification mix, and social inequalities become stabilised with recourse to the supposed visual evidence. But if confronted with the unexpected, we can revise and ‘re-calibrate’ our predictions and perceptions. This is where we imagine the potential for artistic interventions. Art can encourage new ways of ‘seeing’ and shift the boundaries which we apply on otherwise continuous nature of visually accessible objects, including bodies, their shapes and skin tones.

The challenge of the scientific gaze remains. Sciences use categories in multiple and often unreflexive ways and contribute to their fixation. For example, mono-categories are routinely used in experimental psychology to examine stereotyping. Complex concepts such as race are at times reduced to simple categories (resulting in colourism). Categories often are used in simplified additive way, for example in social surveys in sociology, health, or education studies (Bauer et al. 2021). Categories in the sciences are used to denote a mode of perception (to ‘order the world’), a mode of experience, and/or as an analytical lens to explain human perception and/or experience of difference. References to visual, sensory experience and the visually accessible features of objects crosscut each of these usages, contributing to the confusion rather than resolving it.

The challenge is not the lack of exchange between critical theory and cognitive science—we can think of the cognitive turn in sociology starting with Habermas, the attempts to integrate the socio-cultural and naturalistic approaches in cognitive sociology, or the influence of critical theory in psychology leading to the development of critical psychology (cf. Teo 2014; Strydom 2019)—but rather the routine application of categories of race or gender in empirical and experimental studies. In this sense, there is perhaps sufficient mutual influence between critical theory and the sciences on the level of theory, but insufficient transformation regarding the methodologies. It requires thus a change of scientific culture to enable more intersectional seeing.