Social psychological theory and research on the novel coronavirus disease (COVID‐19) pandemic: Introduction to the rapid response special section

Our quest is not to find the one truth or “the science,” but to deploy our different academic insights to best effect under the circumstances. […] That is what we need to be doing now by drawing on every sphere of academic insight across the humanities, social and natural sciences. If we can get that right and stick with it, we will lay the foundations for shaping a better future (Abrams, 2020).

Viral transmission is dependent on human behaviour. Slowing the transmission of COVID‐19 has required people globally to undertake significant and profound behavioural changes almost overnight – and continue to comply with these changes. How to reduce viral transmission is therefore as much a question of social psychology as it is of virology and epidemiology, and requires the careful deployment of all we know about the factors that influence collective solidarity and lasting behaviour change. This requires ongoing work and detailed empirical analysis, but it also necessitates an immediate response to articulate how the global effort to slow the spread of the virus can be informed by what we already know about social behaviour. The contributions to this special section therefore represent an initial endeavour to apply the lessons of social psychological theory and research to the pandemic, in the hope that these insights can directly inform public policy, as well as individual and collective behaviour.

The considered application of pertinent and rigorous social psychological theory and research to the pandemic is particularly important because the responses of many governments appear at times to have been informed by flawed psychological assumptions based around notions of panic, behavioural fatigue, and psychological frailty (Drury, Reicher, & Stott, 2020). By contrast, as the contributions to the special section make clear, public reactions to the pandemic have been contingent on trust, values, leadership, perceptions of personal and collective efficacy, collective identity, and social norms (the latter of which can be both helpful and unhelpful) (see Van Bavel et al., 2020; for a review).

Introduction to special section papers

There is very little published behavioural science research on pandemics, and therefore, this special section represents the first collection of peer‐reviewed theoretical papers and empirical studies that are dedicated to increasing our scientific community's insight into social psychological aspects of global public health. The pandemic raises myriad social psychological challenges and questions, and the papers presented here aim to elucidate just a few key issues. Below, we note the themes across these papers. Each special section paper makes practical suggestions for public policy and to increase facilitative (and decrease unhelpful) behaviours to reduce or limit viral transmission and minimize the negative social psychological consequences of COVID‐19. We summarize the key contributions in Table 1.

Table 1.

Overview of brief reports in special section

Brief report	Author(s)	Key conclusions	Practical recommendations
Theme 1: The role of pre‐existing identities, norms, values, and worldviews in determining responses to the pandemic
1.	Cruwys et al. (2020)	Existing approaches to modelling virus transmission inaccurately model human interaction. People who share social identities perceive each other as less risky and therefore undertake greater risk‐taking behaviour in in‐group interactions because they experience more trust and less disgust.	Mathematical models of transmission should weight interactions by psychological group membership (social identities), because group membership moderates perception of health risk and in turn affects health risk behaviour. Public health messaging should take into account the moderating role of shared social identities on compliance with physical distancing advice. When public health campaigns highlight the importance of physical distancing, the campaign should simultaneously aim to communicate that physical distance is an expression of in‐group care, and explain that this includes proximal social connections.
2.	Van Assche et al. (2020)	Experimental evidence (N = 377) suggests that salience of outgroup non‐compliance with public health guidance increases negative moral emotions, and, in turn, support for retributive measures.	Making salient intergroup differences in compliance with public health advice could increase intergroup tensions, including discrimination. Therefore, governments and officials should avoid framing normative conduct along group boundaries. Instead, communications should be inclusive and aim to promote togetherness, collective resilience, and solidarity across group boundaries.
3.	Courtney et al. (2020)	Drawing on the terror management health model (TMHM; Goldenberg & Arndt, 2008), Courtney et al. (2020) argue that distal defences to threats to health, involving deriving symbolic meaning from cultural frameworks, occur automatically to combat non‐conscious accessibility of death thoughts. This can increase intergroup bias if such bias is intrinsic to pre‐existing values and cultural frameworks. People's decisions about health behaviours depend upon the alignment of these behaviours with their values, cultural frameworks, and worldviews.	Compliance with public health advice is most likely when a behaviour is perceived as easy, immediately actionable, and effective for reducing the threat. Communications should aim to resonate with worldviews that support collectivism and taking collective responsibility to increase compliance with public health advice.
4.	Wolf et al. (2020)	There are likely to be higher levels of pro‐social behaviour, and of compliance with restrictions on movement, amongst people who are high in transcendence and conservation values. Evidence indicates that persuasive communication can be designed to take account of these values to maximize the likelihood that those who don't subscribe to them will engage in the appropriate behaviours. Perception of value similarity is also key.	Tailor communications to take values into account and ensure that value mis‐perception is corrected. For example, highlight high levels of compliance with restrictions rather than occasional non‐compliance.
Theme 2: The emergence of new groups and norms in reaction to the pandemic
5.	Jolley and Paterson (2020)	A cross‐sectional survey (N = 601) provided evidence that belief in conspiracy theories is related to increases in anger, which in turn interacts with an individual's level of paranoia to predict increases in violent action intentions and justification of violence.	Interventions should target and teach paranoid individuals to respond to the anger they feel in response to conspiracy beliefs in non‐violent ways.
6.	Maher et al. (2020)	Shared opinions about the pandemic can become the basis of social identities, and as such provide a psychological platform for compliance/ non‐compliance with public health advice. A 3‐wave longitudinal analysis (N = 253) of public health attitudes and a follow‐up survey (N = 217) showed that distinct opinion‐based groups formed and diverged over time, partially reflecting pre‐existing societal divides, and which predicted compliance with public health advice.	Public health messages should be tailored and targeted to groups who trust/do not trust science to maximize behavioural compliance and avoid intergroup polarization.
7.	Prosser et al. (2020)	As lockdown measures are eased there is potential for new identity dynamics to develop. Specifically, it may be that those who continue to abide by the regulations are seen as moralizing, and thereby become subject to ‘do‐gooder derogation’.	Individuals should adopt a cautious approach to norm negotiation in inter‐personal contexts. Policymakers should consider the role of moral content in persuasive communications, and consider maintaining institutional messages regarding what is expected (e.g., retaining distance‐markers on shop floors). Care should be taken to avoid inadvertently communicating the message that social distancing (and similar) measures are no longer required as lockdown is eased.
Theme 3: How to effectively collectivize responses to the pandemic
8.	Biddlestone et al. (2020)	Evidence from a cross‐sectional survey (N = 724) suggests that horizontal collectivism is positively related to compliance with public health advice because it is associated with decreases in feelings of powerlessness (and thus people feel that their behaviour can have a meaningful impact on viral transmission). Furthermore, vertical individualism was negatively related to physical distancing intentions, both directly and indirectly through its association with increases in belief in conspiracy theories and increases in powerlessness.	Promoting collectivism (‘we are all in this together’) may increase compliance with physical distancing and hygiene behaviours, thus improving public efforts to reduce viral transmission.
9.	Templeton et al. (2020)	Threats are not equal: Social inequalities affect people's ability to comply with public health advice. Non‐compliance can, in turn, decrease community cohesion and increase discrimination, and worsen the long‐term implications of the pandemic.	Governments should facilitate equal ability to comply with the new public health advice, by respecting the different needs of different communities, and should prioritize accounting for structural inequalities in policy and guidance to avoid alienating vulnerable groups and preventing them from being able to follow the advice. Post‐pandemic, additional support should be provided to those groups who were disproportionately affected by the pandemic.
10.	Drury et al. (2020)	Provides a three‐fold classification of explanations for deaths in disasters as a counter argument to explanations that are based on notions of collective panic, selfishness and psychological frailty. Chiding people as ‘selfish’ and for ‘panicking’ individualizes the problem and increases the likelihood of selfish behaviour. In reality, people tend to under‐react to threat, but systemic factors mean that disasters do not affect everyone in the same way, leaving some people with fewer behavioural choices.	Official messaging should build trust by treating the public with respect and openly communicating public health information, and aim to collectivize and normalize – rather than individualize – the risks and effects of (not) adhering to public health advice.
11.	Reicher and Stott (2020)	Whether the COVID‐19 pandemic leads to order or disorder is likely to be influenced by the historical context of relationships between the people and the state, the nature of political leadership within the pandemic and the perception and reality of procedural justice. Additionally, it is important that policing measures proceed from the assumption of shared identity, rather than risk reinforcing division.	The immediate behavioural responses to the pandemic cannot be divorced from longer‐term social trends, or from deep‐rooted social inequalities. Long‐term practical initiatives to address structural inequality are needed. Policing should be geared more towards enablement than enforcement.
12.	Elcheroth and Drury (2020)	The evidence base on how people behave in crisis situations highlights the importance of taking into account the malleability of social behaviour, the role of social identity and collective continuity, and the need for clear communication and practicable guidance and regulations.	Policy makers should anticipate rapid social change and make constructive behaviour visible in order to ensure this change occurs in desirable directions. Guidance should be clear and easy to put into practice, and inclusive role models should be used in disseminating this guidance. Encouraging recollection of past ordeals and how these were overcome will help build solidarity, and care should be taken to avoid reinforcing the myth of collective panic. Emergent communities should be allowed to develop organically, and efforts should be made to demonstrate shared vulnerability and to allow for the continuity of social ties.

This article is being made freely available through PubMed Central as part of the COVID-19 public health emergency response. It can be used for unrestricted research re-use and analysis in any form or by any means with acknowledgement of the original source, for the duration of the public health emergency.

The role of pre‐existing identities, norms, values, and worldviews in determining responses to the pandemic

The first theme that permeates the reports in this special section is the role of pre‐existing group identities, norms, values, and worldviews in affecting both people's (non‐)compliance with public health advice, and how they respond to in‐group and outgroup members differently during the pandemic. Cruwys, Stevens, and Greenaway (2020) argue that people who share group membership (a social identity) feel less vulnerable to disease transmission from each other during interactions. In‐group members – the people we trust most, and interact with the most, can pose a greater risk, in terms of viral transmission, to ourselves than strangers (see also Hult Khazaie & Khan, 2019). The key moderator of the positive impact of shared group membership on health behaviours is the nature of in‐group norms (see Cruwys et al., 2020; Templeton, Guven, Hoerst, Vestergren, Davidson, Ballentyne et al., 2020). The implication is that the mathematical models and contact tracing mechanisms that are being used to predict and limit the spread of transmission of the virus should weight social contact and connections according to the social identities of the individuals within the network, and the associated in‐group norms. To achieve a more accurate prediction of viral transmission, it is vital that future research explores the viability and logistics of weighting social interactions in mathematical models of viral transmission in this way.

Whilst it is in‐group members and in‐group norms that may be most influential in affecting (non‐)compliance with the recommended public health behaviours, Van Assche, Politi, Van Dessel and Phalet's (2020) data suggest that it is salience of outgroup non‐compliance with public health guidance which increases negative moral emotions, and which in turn increases support for retributive measures. This suggests that making salient pre‐existing group boundaries and differences between the behaviour of groups can potentially increase discrimination (see also Templeton et al., 2020). This is echoed by the international rise in prejudice and discrimination against Asian people (Wen, Aston, Liu, & Ying, 2020), and in the fact that right‐wing groups and individuals are using the pandemic to fan the flames of xenophobia (Bieber, 2020). This means that the pandemic has exacerbated pre‐existing societal divisions and biases already, in terms of increasing racism and intergroup conflict, and will continue to do so unless appropriate mitigating strategies are put in place to support and protect those groups most vulnerable to discrimination (Templeton et al., 2020), and shape narratives as inclusive rather than divisive. If governments do not take action to mitigate the interaction between social inequalities and the impacts of COVID‐19, discrimination may be enhanced by the very responses and guidelines for public behaviour that should function to unite people (Templeton et al., 2020).

Courtney, Goldenberg, and Boyd (2020) suggest that the increase in prejudice and discrimination may be at least partly explained by a distal reaction to the threat of COVID‐19. According to terror management theory (Greenberg, Pyszczynski, & Solomon, 1986; Greenberg et al., 1990), people's proximal reactions to health threats may involve denial, suppression, or immediate actions to reduce the threat. Then, once threat perceptions are sufficiently reduced (or if conscious threat perception was bypassed), death concerns can become non‐conscious. It is during this phase that people lean into their worldviews. If those worldviews promote cultural values of self‐enhancement or ideologies like nationalism, this has the potential to increase xenophobia, racism, and other forms of intergroup discrimination and outgroup derogation. The opposite is also true: Specific worldviews can also promote values such as self‐transcendence and a decrease in intergroup conflicts (see Wolf, Haddock, Manstead, & Maio, 2020).

People's decisions about health behaviours depend upon the alignment of these behaviours with their pre‐existing values, cultural frameworks, and worldviews (Courtney et al., 2020). Indeed, Wolf et al. (2020) highlight the role of pre‐existing values in the way in which people respond to public health measures introduced by governments to stop the spread of the virus. Drawing in particular on Schwartz's (1992) quasi‐circumplex model of values, they argue that evidence suggests that people who hold socially oriented values of self‐transcendence and conservation would be likely to comply with public health guidelines and regulations, and to engage in greater levels of pro‐social behaviour. By contrast, those who hold more personally focused values of self‐enhancement and openness might be expected to be less likely to comply and to engage in pro‐social behaviours. However, importantly Wolf et al. (2020) note that these values need not be thought of as rigidly constraining action, and they argue that persuasive communications can be formulated in such a way as to elicit greater levels of compliance and pro‐social behaviour even amongst individuals whose values may not pre‐dispose them towards it. Of particular note, they argue that it is crucial to shape communications in such a way as to encourage people to see how the majority of others are abiding by the guidelines, rather than to draw attention to a minority who may be flouting them.

The emergence of new groups and norms in reaction to the pandemic

Whilst the first four brief reports in this special section underscore the significant role of pre‐existing groups, values, norms, and worldviews, the pandemic also has the capacity to exacerbate and create new societal divisions. There are three brief reports that provide insights into these processes. First, Jolley and Paterson (2020) describe when and why violent groups have emerged during the pandemic. Across the world, people have violently targeted 5G engineers and telecommunication masts due to conspiracy theories that implicate 5G technology in viral transmission. Jolley and Paterson (2020) suggest that belief in such conspiracies can lead to violence when increases in anger interact with individuals’ paranoia.

The groups that have vandalized telecommunication masts may represent new opinion‐based activist groups (see Bliuc, McGarty, Reynolds, & Muntele, 2007; McGarty, Bliuc, Thomas, & Bongiorno, 2009). Maher, MacCarron, and Quayle (2020) demonstrate that new psychological groups can form based on shared attitudes towards the pandemic and towards outgroups that are made relevant and salient by the pandemic. These new opinion‐based groups may build upon the pre‐existing groups we discussed above, but are not reducible to these pre‐existing groups. As new groups form and change, they in turn provide the basis for compliance or non‐compliance with public health advice (Maher et al., 2020). This is likely to be a recursive process: Public health communications and the actions of officials can divide opinion and lead to the formation of new opinion‐based groups, and when opinion‐based groups form, they are likely to impact on how people perceive and understand, and comply with, public health advice (Maher et al., 2020).

Prosser, Judge, Bolderdijk, Blackwood, and Kurz (2020) suggest that as the pandemic continues and initial strict lockdown measures and other social restrictions are eased, continued adherence to guidance around practices such as handwashing and physical distancing may become a moral issue, and moreover that some people might find themselves the subject of exclusionary sentiment on the grounds of sticking rigidly to the guidelines. This phenomenon of ‘do‐gooder derogation’ (Minson & Monin, 2012) highlights the need for governments and public health agencies to be aware of the possible negative unintended consequences of giving a premature impression that things are returning to normal. For example, as different territories move through various stages of easing lockdown restrictions, it becomes ever more important to highlight the need to maintain physical distancing to continue to slow the spread of the virus.

How to effectively collectivize responses to the pandemic

The brief reports described above suggest the boundary conditions for when collectivizing a threat can backfire to produce negative societal consequences. Yet, the final five brief reports in this special section argue that governments and public health campaigns should aim to leverage the power of the people and collectivize people's responses to the pandemic. In promoting a collective response to the pandemic, governments and officials should aim to normalize – rather than individualize – the risks and effects of (not) adhering to public health advice. As Drury et al. (2020) argue (see also van Bavel et al., 2020), when politicians frame people's behaviour in individual terms, for example by focusing on individuals washing their hands or buying too much toilet paper, they individualize people's response to the threat. This, in turn, functions to individualize future responses, potentially creating the very problem (i.e., ‘panic’ buying, belief in conspiracies, see Biddlestone, Green, & Douglas, 2020; Jolley & Paterson, 2020) that the communications aimed to prevent. Instead, by emphasizing the shared self‐relevance of the threat and the efficacy of a united and inclusive collective response, governments can leverage the psychological power of shared social identities to change behaviour and sustain that change (see Smith, Thomas, & McGarty, 2015).

Collectivization should be done whilst keeping the aforementioned boundary conditions (pre‐existing identities, norms, values, worldviews) in mind. For example, Biddlestone et al.'s (2020) study suggests that people with a more collectivist orientation have greater intentions to engage in hygiene‐related behaviours; therefore, promoting collectivism may facilitate engagement with public health advice. Courtney et al.'s (2020) analysis also supports the need to shift the focus and responsibility from individuals to collectives. However, careful attention should be paid to in‐group norms of those collectives: Cruwys et al.'s (2020) arguments suggest that when public health campaigns try to collectivize a threat and the response to that threat, the campaign should simultaneously target the formation of an in‐group norm of physical distance as an expression of in‐group care.

Such collectivization strategies also need to be multi‐layered. Due to deep pre‐existing structural inequalities, governments need to protect and support groups in the community who are most vulnerable to COVID‐19 and also the least able to comply with the new norms based on public health advice. This protection should be embedded not only in addressing the very roots of those inequalities but also in proactively facilitating the ability of members of those groups in being able to comply. The consequences of such differentiation in public health policy would go beyond affecting viral transmission; it would protect vulnerable groups from the social consequences of non‐compliance (Templeton et al., 2020). Indeed, as Reicher and Stott (2020) point out, without an understanding of the broader structural problems and inequalities that underpin the varying responses to the pandemic, we may be condemned to repeat the mistakes of the past. There have undoubtedly been a great many displays of collective solidarity around the world, with communities coming together to sing, dance, and applaud, all whilst maintaining appropriate social distancing. But if these expressions of solidarity are not coupled with concrete action and change, then we may find we are no better prepared for similar future emergencies. To give just one example, in the UK the ‘clap for carers’, in which people applauded health care workers every Thursday evening at 8 pm for the first 10 weeks of lockdown has rightly been highlighted as an example of the emergence of an (initially spontaneous) manifestation of collective gratitude and support. However, as time wore on questions began to be asked both of its increasingly moralized normative nature (i.e., people perhaps began to feel obliged to participate), and – more searchingly – about the contradictions inherent in applauding health care workers whilst engaging in behaviours that either made their job more difficult (e.g., failing to maintain physical distancing in the context of street parties arranged to mark the 75th anniversary of VE Day; BBC News, 2020), or supporting government policies that have seen increasing pressure on the funding of the National Health Service (Anonymous, 2020). To frame this positively, the solidarity that has emerged around health care workers represents a unique opportunity to mobilize people to ensure that the polite applause is followed by united community action and sustained public investment.

Reicher and Stott (2020) also highlight the role of political leadership, procedural justice, and policing in building collective identity, whilst warning of the risks of creating division and distrust if those people who shape the policies and guidelines on the pandemic fail to adhere to those policies and guidelines (see also Templeton et al., 2020).

Whilst Reicher and Stott's (2020) arguments highlight that there are longer‐term lessons to be learned from the current crisis, Elcheroth and Drury (2020) draw on the literature on behaviour in emergencies to underscore how a number of more immediate practical implications follow from the same broader tradition of social identity theorizing. Crystallizing some of the messages highlighted in earlier papers, Elcheroth and Drury's (2020) ten key recommendations focus on the way in which communication strategies can be designed to effectively take into account the potential to build a pro‐social sense of collective identity and to show people that fairness and procedural justice are at the heart of the response to the virus.

A note on the implications of the pandemic for diversity in research

In setting out the lessons of social psychological theory and research for the response to the pandemic, we are acutely aware that our response as a disciplinary community is similarly impacted by matters of values, identities, norms, and structural disadvantage, and that – as with the population more broadly – some social psychologists, and some specific groups of social psychologists, may be more affected by the current circumstances than others. With school closures, illness, and caring for loved ones, this crisis has disproportionately impacted some of our authors and reviewers (Minello, 2020; Viglione, 2020). To explore this systematically – albeit in a necessarily preliminary and post‐hoc fashion – we conducted an analysis of the demographic diversity of our submitting authors. On the 108 submissions, we received for the special section; there were 387 authors, of whom 55% were male. The frequency of male and female authors differed significantly with seniority, with a higher proportion of male authors in senior positions than female authors in senior positions, χ² (2, N = 387) = 32.87, p < .001. The submitted manuscripts were not more likely to be written by a male lead author than a female lead author, but male lead authors were more likely to be in a senior role than female lead authors, χ² (2, N = 108) = 7.92, p = .02. On those papers on which the lead author was female, these authors tended to be more early‐career than their male counterparts, indicating that there was a gap in the diversity of our submitting authors: female, mid‐career academics. Clearly, a fuller analysis would necessitate controlling for the proportion of male and female academics in various roles across the field as a whole, but at the very least this trend appears to support the analyses and predictions of Minello (2020) and Viglione (2020) at the start of the pandemic. However, although there were significant differences in these aspects in the submitted manuscripts, there was no relationship between author sex, χ² (2, N = 387) = 0.40, p = .82, or seniority, χ² (2, N = 387) = 3.77, p = .15, and being accepted for publication in this special section (there was also no relationship between author sex and publication outcome or seniority and publication outcome when analysing the data for lead authors only: author sex, χ² (2, N = 108) = 0.85, p = .65; seniority, χ² (2, N = 108) = 5.05, p = .08). Nevertheless, we offer these initial exploratory analyses not as an attempt at the final word on the matter, but rather as a stimulus to further consideration of these issues. We welcome further analyses on the theme of author diversity and on the disproportionate impact of COVID‐19 on specific groups in academia, as in all other contexts.

Future research directions

Clearly, the COVID‐19 pandemic is a rapidly developing situation and the contributions collected in the present special section constitute only an initial indication of the ways in which social psychology might draw on its stock of established theory and research evidence in order to suggest immediate ways forward. Of equal importance are the possible future directions of research on the social psychological aspects of the pandemic, and these will need to be developed to take into account the changing social context. One such development that emerged after the bulk of the contributions to the special section had been completed is the collective protests and action in locations across the world in response to the murder of George Floyd in Minneapolis, and which have rallied around the wider Black Lives Matter movement. The emergence of this collective action in the midst of the pandemic awaits a fuller analysis, but a number of our contributors make reference to the danger of (presumably) well‐intentioned measures and communications designed to address the pandemic inadvertently reinforcing existing social divisions and inequalities. This must clearly be a key focus for research in the months and years ahead.

There is a further need for longitudinal research mapping the changing trends in the social psychological dynamics of the pandemic. Indeed, if Prosser et al. (2020) and Maher et al. (2020) are correct, then there is scope to explore how new groups and identities emerge as a result of these very specific circumstances. The group processes involved will no doubt reflect some well‐established theoretical principles, but equally we should be alive to the need to modify our theories and models in response to the emergent behaviours we see in response to the continuation of the pandemic in the coming months.

A tradition of work under‐represented in the special section is the careful analysis of talk and text, which has become an integral part of social psychology in recent decades (e.g., Potter & Wetherell, 1987; Wiggins, 2017) The potential to explore the delicate choreography of social distancing, the changed norms of interaction, and the rhetoric of political communication remains to be fully explored. Indeed, as Reicher and Stott (2020) note when discussing the role of historical myths in the response to the crisis, we should beware reifying these social constructs, for they are ‘not so much determining cultural presences as rhetorical resources that can be drawn upon – and indeed need to be actively invoked – if they are to influence our understanding of COVID‐19, our relationship to authority during the pandemic and our responses to their policies’ (pp. 3–4).

Conclusion

This rapid response special section is intended to be the start, rather than the conclusion, of a conversation about how behavioural science can inform policy and behaviour around the pandemic. The brief reports in this special section illustrate the wealth of conceptual resources that social psychology can draw on to inform individual, group, and state responses. As this is just the beginning of academic endeavour related to COVID‐19, we want to encourage further submission of responses to, and empirical work informed by, these special section reports in the hope of continuing the conversation in the pages of BJSP. As we write, it is still highly uncertain how the trajectory of the pandemic will unfold, and what its immediate and more distal social consequences might be. It is certain, however, that if the medical and epidemiological response to this crisis continues to be mirrored by an equally vigorous social psychological response, then we may be able to move beyond the current situation united by our common humanity, better prepared for future pandemics, and better placed to address the long‐term structural inequalities that have been so cruelly laid bare and exacerbated by COVID‐19.

Author contributions

Laura G. E. Smith, PhD (Conceptualization; Formal analysis; Investigation; Writing – original draft; Writing – review & editing) Stephen Gibson (Conceptualization; Writing – original draft; Writing – review & editing).

Data availability statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.