Enacting the ‘consuming’ brain: An ethnographic study of accountability redistributions in neuromarketing practices

Introduction: The rise of the brain

The figure of the brain has risen to prominence in science and society over a period of at least 30 years, and shows little current sign of abating. In the context of the development of sophisticated neurotechnologies, notably the visualisations afforded by functional Magnetic Resonance Imaging (fMRI) brain scanning, neurosciences have become an important part of the psychological sciences (Rose & Abi-Rached, 2013). President George Bush Senior formally designated the period 1990–2000 ‘the decade of the brain’, ¹ since when neuroscience research effort, funding and resources have continued to grow as evidenced by, for instance, the United States’ National Science Foundation’s (NSF) goal directed at ‘Understanding the Brain’, including a multi-year activity that includes NSF’s participation in the ‘Brain Research through Advancing Innovative Neurotechnologies’ (BRAIN) initiative as well as the scheme ‘Next Generation Networks for Neuroscience’ (NeuroNex) to establish global research networks of excellence. Similar initiatives in Europe include the European Union’s Horizon 2020 Research and Innovation Programme for the Human Brain Project. At the same time the position of the brain in popular culture and social and political discourse continues to solidify (Broer & Pickersgill, 2015; Thornton, 2011). In these settings the brain is frequently considered as the acknowledged centre of cognition, knowledge and affect (Rose & Abi-Rached, 2013). ‘Neuro’ is the epithet of choice to redesignate a range of disciplines and activities: we have seen the emergence of neuroeconomics, neuropolitics even neuro kitchen design. ‘Brain training’ apps enjoy widespread popularity. ² A large array of recent consumer products feed the notion that the healthy functioning of the brain is enhanced by particular food and drink products: one brand offers variants promising to enhance a wide range of life’s concerns and activities including fitness, sleep and more. ³ Sociologists Rose and Abi-Rached (2013) assert that: ‘it is not that human beings are brains, but we have brains. And it is in this form – that our selves are shaped by our brains but can also shape those – that neuroscientific arguments are affecting conceptions of personhood and practices of self-fashioning’ (Rose & Abi-Rached, 2013, p. 22).

So the rise of the brain is associated with wide-ranging consequences and is said to give rise to profound questions about our understanding of human behaviour (Choudury & Slaby, 2012; Vidal, 2009). In this article we ask, to what extent and in what ways does the brain supplant the person as the presumed origin of human behaviour as claimed by neuromarketers, who study consumers’ preferences? And what are the implications of this claim in relation to neuromarketing research practices? Explanations of behaviour centring on the brain rather than the person suggest a shift in understandings of accountability. In order to get to grips with these concerns, we argue that more granularity is needed. We need to look in detail at how questions about the claimed move from person to brain map onto, or derive from, instantiations of the brain in practice. To investigate this, we specifically focus on studying shifts in accountability between person and brain.

Our study is based on a multi-sited ethnography of the experimental articulation of the consuming brain in the growing field of neuromarketing. Fisher et al. (2010) see neuromarketing as ‘one manifestation of [a] new neuroculture’ (2010, p. 230). Those working in the field describe neuromarketing as a ‘hybrid field’ (Plassmann et al., 2012, p. 18) that purports to adapt a number of methods and theories from neuroscience to understand consumer behaviour. Typically, this involves the use of brain imaging, scanning, or other brain activity measurement technology as well as biometric methods, to capture consumers’ brain responses to marketing stimuli. ⁴ Proponents of consumer neuroscience (as researchers in academia, mostly, marketing, refer to the field) and neuromarketing (the term commonly employed by market research professionals and in the media) promise that the use of neuroscientific and biometric technologies provides consumer insights that are more accurate than those generated by other methods such as surveys and focus groups (Boksem & Smidts, 2014; Karmarkar & Plassmann, 2019). Neuromarketing companies characteristically describe consumers as not able to explain their consumer behaviour (Morin, 2011; Schneider & Woolgar, 2012) and hence distinguish themselves from traditional market research, which primarily uses questionnaires or focus groups to determine consumer insights. Similarly, academic researchers in consumer neuroscience emphasise the unique insights neurotechnologies provide compared to other methods (Schneider & Woolgar, 2012). In other words, both groups claim that neuroscientific and biometric measures enable the determination and even prediction of unbiased and internal consumer responses to marketing stimuli. The key to understanding consumer behaviour accordingly lies in understanding consumers’ brains, as is the opinion of these two groups.

We suggest that this development towards ‘brain-based’ explanations of human behaviour (in the sphere of consumption) represents a more general tendency to explain human behaviour by relying on what is sometimes referred to as ‘hard sciences’ (e.g. biology, neuroscience, data science). Our article emphasises that the explanatory power of these purportedly more objective sciences and approaches rests on the uptake and employment of novel technological artefacts and/or procedures that are considered neutral. Such a view is highly contested in sociology, anthropology, Science and Technology Studies (STS) and beyond (Dumit, 1999; Porter, 1995), yet these debates have resulted in limited interdisciplinary or public discussion of and reflection on technological optimism, the belief that technologies will (neutrally) advance knowledge and contribute to national and economic development.

In this article, we report our research into neuromarketing experiments. The focus of our research attention is to examine the shifting attention from (purportedly) subjective knowledge of the consumer to (purportedly) objective knowledge of the brain during the neuromarketing study. The article is divided into four sections. We begin by reviewing existing sociological and anthropological research on how the brain features as locus of objectivity in the neurosciences. Secondly, we present our conceptual framework drawing on recent contributions in the field of Science and Technology Studies (STS) to suggest an alternative to the constructed dichotomy between objective and subjective knowledge. We suggest that beyond studying the representation of the brain we need to understand the processes of enactment of the brain(s) in situated practices (Mol, 2002) and how this affects accountability relations (Woolgar & Neyland, 2013). Thirdly, we describe our multi-sited ethnographic approach to neuromarketing in Europe. We discuss three fieldwork vignettes in detail, paying particular attention to how the ‘consuming’ brain is enacted in neuromarketing practices. We show how ontological respecification of the consumer is accomplished in experimental practices and how this redistributes accountability relations between person and brain. We conclude by emphasising the contributions our findings make to situate the wider rise of brain-based and other technologically-enabled explanations of human behaviour and call for studying accountability in practice.

Sociological and anthropological engagements with the neurosciences: From personhood to brainhood

Rose and Abi-Rached (2013) assert that the brain sciences are transforming the ways we ‘know ourselves’ as human beings. Vidal (2009) argues that brainhood rather than personhood is gaining prevalence in the neurosciences and in public discourse. Yet, studies by, amongst others, O’Connor et al. (2012) complicate this picture. In their study of British news media they show that despite reportage about neuroscience that depicted the brain as ‘biological proof for the legitimacy of particular phenomena and beliefs’ (p. 221, emphasis in original), the brain was also frequently portrayed as a resource to be optimised and ‘an index of difference’ to draw boundaries between different categories of people. Ultimately, the authors show that a rise of the neurosciences has not led to revolutionary understandings of personhood among publics in the United Kingdom. Other researchers have also relativised a radical shift from personhood to brainhood when studying the impact of the neurological gaze on subjects such as patients, children, or teenagers (Bröer & Heerings, 2013; Choudhury et al., 2012; O’Connor & Joffe, 2013; Pickersgill et al., 2011). According to these studies, people who are confronted with their biological constitution do not simply adopt a reductionist perspective of their beings, but reconstruct their own ideas of agency with the knowledge that is provided (Pickersgill et al., 2011). That is, people sometimes refer to their brain or neurochemicals as the actor of their behaviour, but at other moments they refer to themselves, their willpower, their emotions, or their social situation, as the cause of their actions (e.g. Brenninkmeijer, 2016). Moreover, some studies even demonstrated that neuroscientists rarely seem to perceive themselves as defined by the brain (Pickersgill et al., 2011).

Interestingly, neuroscientific experiments are not exclusively focused on the brain either, as ‘laboratory studies’ ⁵ of neuroscience suggest. Mahfoud (2014) emphasises in her review of extant ethnographies of neuroscience laboratories in the United States and Europe (e.g. Alać, 2008; Beaulieu, 2004; Roepstorff, 2001) how these studies reveal the role of interactions between bodies, artefacts and their environment in the experimental setting and how these co-constitute knowledge in neuroscience laboratories. For instance, Mahfoud (2014) reviews how the anthropologist Roepstorff (2002) ⁶ describes the transformation of subjects into objectivity based on his ethnographic study of a brain imaging laboratory in London. His account shows how subjects participating in the brain imaging studies are turned into data and the role researchers, the place/setting, brain mapping software and, thus, the interactions between people and artefacts play in the process. He compares, for instance, the house in which the laboratory is located to ‘a black box, where subjects enter at the bottom and objectivity exits at the top’ (Roepstorff, 2002, p. 154 as cited in Mahfoud, 2014, p. 3).

However, notions of objectivity in neuroscience change due to the introduction of digital tools and resources, as Beaulieu (2001) highlights in her ethnographic study of brain atlases developed in the 1990s US Human Brain Project. Considering atlases as ‘repositories and enforcers of objectivity’ (2001, p. 638), she argues that we can see the formation of ‘digital objectivity’ that enables ‘the bypassing [of] human judgement in the evaluation of large numbers of scans’ (2001, p. 667). She emphasises that:

. . . the displacement of expertise away from an embodied observer has non-trivial consequences for notions of objectivity. If judgement and experience are delegated to technologies in neuroinformatics, it is likely that other notions that have been part of scientific practice and traditionally located in the subject (creativity, understanding, discovery and insight) will also be reshaped by informatics. (Beaulieu, 2001, p. 669)

Beaulieu emphasises that the shift of expertise away from human observer to technology classifies the technology as more objective. It is this shift and what it implies for the making of objectivity, that we ethnographically explore in our study of neuromarketing experiments.

Yet, the person is not completely absent. In an ethnographic study of two fMRI laboratories, Alač (2008) shows that making sense of brain scans is not merely a process of looking at the data, but includes gestural engagement of the scientists: ‘Gesture, talk and the manipulation of the digital screen function together as techniques for managing perception’ (2008, p. 493). Research by Cohn (2008) demonstrates that neuroscientists tinker with their experimental settings, their subjects, and their statistical results to constitute neuroimaging results. This is also the case in neuromarketing experiments as we have shown in a recent publication (Brenninkmeijer et al., 2020). A study of a neuroscientific therapy demonstrated that the experimenter had to work with ‘the whole person’ not only with ‘the brain’, for example by calming down people’s minds, or correcting their bodies. (Brenninkmeijer, 2016, p. 127). And an anthropologist who became a neuro test-subject was told that it ‘doesn’t matter’ that she could see her own brain responses during the experiment, and expressed her surprise at this ‘lack of interest in subjectivity’ in such experiments (Martin, 2013, p. 149). That made us wonder: if patients, test-subjects and neuroscientists do not see themselves and each other as brain primarily or exclusively, and if neuroscientific experiments are not as simple and effortless as their reported outcomes imply, how then are consumer brains enacted in practice?

Brains in practice: Ontology and accountability

STS provides important analytical concepts and methodological procedures for studying how the brain is rendered accountable for consumer behaviour. By attending to ‘techniques that make things visible, audible, tangible, knowable’ (Mol, 2002, p. 33), we can observe how an entity is enacted in practice. In her ethnography of medical practices in a Dutch hospital Mol (2002) has argued for a shift from epistemology to ontology, i.e. from examining which representation of reality is accurate to considering how practices enact reality differently. Foregrounding the multiplicity of entities (objects, technologies, persons) and their practice-based accomplishment complicates the distinction between subjective and objective accounts of a phenomenon. This so-called ‘ontological turn’ in STS (e.g. Woolgar & Lezaun, 2013, 2015) and the question of how entities are enacted has spurred an important debate in STS (e.g. Lynch, 2013; Sismondo, 2015) and beyond. Instead of talking about beliefs, representations or culture, researchers explore ‘ontologies’ as entities or objects that have been brought into being and could have been otherwise. Entities as various as salmon or diets can have multiple realities when they are enacted in different practices (Lien, 2016; Mol, 2013). For instance, de Boer et al.’s (2020) study of Non-Invasive Brain Stimulation (NIBS) in the field of cognitive neuroscience explored how the scientific object of ‘visual attention’ is constituted interactively between scientists involved in the development. The study showed that different technologies used to study visual attention ‘do not so much offer different perspectives on one singular object that can be identified as “visual attention”, but rather make scientists relate to different provisional objects that need to be actively stabilized’ (de Boer et al., 2020, p. 2).

Whereas these approaches help us understand how the ontology of entities is achieved, they say rather less about ontological respecification. Our interest is critically to assess the claimed shift from person to brain through an empirically oriented focus on ontological enactment and respecification. By ontological respecification we mean the process whereby the ontology of an entity, what an entity is, is reclassified. We draw on Woolgar and Neyland (2013), who analyse how mundane objects or technologies like bin bags, traffic lights, or ID cards can be subject to an ‘ontological respecification’ when they are made to speak for politics and how this can also change accountability relations. They give the example of a cup of coffee that changes (is ‘respecified’) from ‘coffee’ into ‘too hot coffee’ after a customer successfully sued McDonald’s because their coffee caused her burns – as reported in the media. The ontological respecification of the coffee entailed corresponding changes in accountability relations. While burns arising from the (mis)handling of coffee were previously a customer’s individual responsibility, burns deemed to result from ‘too hot coffee’ made McDonald’s accountable. In similar vein, they discuss the case of the ‘wrong bin bag’, in which a woman was fined by her local council for using the ‘wrong’ type of black plastic bag for waste disposal. Woolgar and Neyland show how an ordinary black plastic sack comes to be viewed as ‘wrong’. The ontology of the bag is reconstituted in and through its (textual) articulation in the news media (2013, p. 50). They argue that ‘the situated ontology of an entity (the bin bag) made possible its apprehension as a moral matter (“wrong”)’ (Woolgar & Neyland, 2013, p. 55).

Woolgar and Neyland (2013) build on Garfinkel’s ethnomethodological understanding of accountability (cf. Linstead, 2006): ‘Activities whereby members produce and manage settings of organized everyday affairs [being] identical with members’ procedures for making those settings “account-able”’ (Garfinkel, 1967, p. 1). For Garfinkel social order and accountability are produced in and through processes of social interaction. Actions can be rendered accountable either by directly observing these actions or by turning them into an account (e.g. a report) that enables others to become witnesses and to evaluate and understand the reported activities. For instance, Suchman’s (1993) study of airport operations rooms shows how the monitoring and reporting of activities – in this case movements of aircrafts – is rendered accountable. The standardised and centralised recording of all activities in a national computer system makes them accountable to different ‘publics’ across time and space. Records need not only be text-focused reports and database entries. Neyland and Coopmans (2014, p. 1) argue that images can become part of the organised relations of accountability. They show how surveillance images, photographs of traffic violations and mammograms are mobilised and contested as evidence. Attending to how visual accountability is realised in these three cases, they emphasise the ‘hailing’ effects of images. By this they mean that images ‘call for and prefigure a certain kind of response and dispersing of responsibility’ (2014, p. 1).

By attending to ‘accountability in action’ (Neyland & Woolgar, 2002), these and other ethnomethodologically inspired ethnographies of accountability contribute to an understanding of how accountability is generated and sustained. Accordingly, Woolgar and Neyland define accountability as ‘no single thing, fixed in time and place. Instead, accountability admits a range of interpretations in terms of the direction, intent, audience for, and distribution of responsibility’ (2013, p. 58). They explain that ‘[f]or ethnomethodologists, what is accountable is a mundane, pervasive organizing orientation for social action’ (p. 32). They distinguish between accountability as a form of intelligibility that attends to ‘making accounts, accountably available’ (p. 32) and responsibility as accounting for conduct in interaction that is one possible effect of accountability. Woolgar and Neyland call for a focus on shifts in accountability as these shifts have significant consequences for how entities are brought into being and sustained through the realignment of accountability relations (Woolgar & Neyland, 2013, p. 58).

To what extent can this conceptual repertoire contribute a different reading of the prominence of the brain? How far does the power of the brain emerge in virtue of reconfigured sets of accountability relations in experimental practices in neuromarketing? Neurotechnologies such as fMRI and electroencephalography (EEG) render consumers’ measured brain activity observable and transform this information into measurements and then visuals (e.g. brain scans) or texts (e.g. published articles, client reports). In the process, we argue, the brain is enacted as an entity that comes to play a pivotal role in the accountability of consumers’ behaviour according to neuromarketing experts. Whereas commonly accountability for knowing about and reporting on individual consumption behaviour rested with consumers themselves, neuromarketing purports to reveal that consumers’ brains are the ultimate source of information about consumers’ preferences. Moreover, as we show, the enacted consuming brain engenders a redistribution of accountability relations in neuromarketing accounts of consumer behaviour. The consuming brain mediates the network of accountability relations and takes a central role, whereas consumers’ accounts become negligible. Who or what has access to revealing why consumers buy what they buy? We claim that this is central to understanding how the shift from person to brain occurs in practice. In particular, we see that the ontological respecification from person to brain is not so much a reduction as an extension and redistribution of accountability relations between consumers, brains, brain scanning operators and neuromarketing consultants.

Enacting the consuming brain: An ethnographic study of neuromarketing

Building on previous laboratory studies of neuroscience, we participated in and observed commercial and academic neuromarketing practices in different sites in Europe intermittently between 2009 and 2015. The study involved attending neuromarketing conferences (in 2011, 2012 and 2014), public lectures (2014) and courses (2014), participant observation of neuromarketing experiments (2012, 2014) and conducting interviews with 23 people from Europe or the United States (intermittently between 2009 and 2014). Interviewees worked for neuromarketing companies, held academic positions at universities and conducted neuromarketing research, or were involved in related activities (e.g. writing books, organising events) to consumer neuroscience. Researchers conducting neuromarketing experiments employed a range of different neuroscientific and biometric techniques aimed at measuring consumers’ biometric and neuronal responses to products or similar stimuli in the laboratory setting. In the following, we introduce each setting and three exemplary vignettes from our fieldwork in detail.

Conclusion

We encountered a diverse range of neuromarketing practices in the different settings we studied and report in the three vignettes above. The classification of marketing practices as ‘neuro’ does not always require the use of what have been conventionally understood as neuroscientific technologies. Neuro can also be attached to the study of consumer preferences through a range of other research techniques and set-ups (e.g. biometric research, experimental psychology) and by way of inference to existing neuroscientific knowledge and the explication how this is relevant for marketing. This brings up the issue of hype about neuromarketing. It also raises questions to what extent neuromarketing knowledge and technologies are employed to render marketing practices more ‘objective’ and ‘scientific’. The latter was discernible in some of our interviews with neuromarketers. For instance, as one of our interviewees, an entrepreneur, who had co-founded a start-up focusing on eye tracking, a biometric measurement of engagement, put it: sometimes it proved useful (for her and the company) to be associated with the neuromarketing ‘banner’ (field notes, May 2011). ¹⁶ All the neuromarketing professionals we visited, observed or interviewed share the position that consumers do not have access to their true motives. This provided the marketers’ rationale for going beyond consumers’ accounts in order to reveal the hidden causes of consumer behaviour (Schneider & Woolgar, 2015). Technologies such as fMRI and EEG are crucial to achieving this revelation according to neuromarketing professionals. However, this achievement rests on the central assumption that the uptake of these technologies is neutral and, more importantly, that this neutrality is essential to achieving valid knowledge about human behaviour. Consumers are thereby secondarised as an object of research and neuromarketing technology and experts assume accountability for revealing why consumers buy what they buy.

In this article we propose a different interpretation by describing how in experimental practice consumers are rendered non-knowledgeable and how the brain is achieved as the locus of objectivity. We have shown in particular how the interlinked actions of multiple (human and non-human) actors contribute to a shift in accountability relations such that the person becomes respecified as a brain. Ontological respecification is accomplished in these neuro experimental settings. But the complexity of these experimental practices belies the arguments that the brain is a mere reduction of the person. In enacting the consuming brain the person is not so much reduced as reconfigured in terms of a specific series of expectations, rights, responsibilities and assumptions about what counts as an adequate explanation of their behaviour. Ontological respecification is not so much reduction as it is a redistribution of accountability relations. As we showed in our ethnographic sections above, accountability is redistributed from person to brain in situ through people conducting the experiment as well as the technologies employed to understand consumers’ brain activities during the experiment. We call for further studies of ontology and accountability relations in practice, so as better to situate novel explanations of human behaviour and to account for the political effects of technologies on transforming agency.

More than this, we might say, the redistribution of accountability is quintessentially messy. For despite the recurrent appeal to the brain as a way of avoiding the unreliability of subjects’ views, we note that all three of our experimental cases also involved the administration of questionnaires. Consumers’ views (in questionnaires) cannot be relied upon except, it seems, when they are extracted as an integral part of an investigation of the brain. From some lofty philosophical vantage point it would be tempting to denounce this as contradiction. More charitably, perhaps, it suggests the intriguing possibility that ironies and contradictions of this kind are necessary in order to achieve the brain. And if the messy redistribution of accountabilities is key to the achievement of the brain, this may provide a clue to the coexistence of different brains in different neuromarketing practices.

Finally, we should note one particular feature of enacting the consuming brain which invites further exploration. As was suggested at the end of the last section, accountability for knowing the brain can be distributed in many different ways. We have described some of the ways in which brains are enacted in neuromarketing practice. Yet our description of these processes itself proceeds by implying distance between our own brains and this descriptive work. What is happening in our brains when we write/read this text? What happens to the universal brain as it apprehends this argument? And, apart from subjecting ourselves to the same kinds of experiment described in this article, how would we know?