Self‐conscious emotions and breastfeeding support: A focused synthesis of UK qualitative research

Abstract Research on women's experiences of infant feeding and related moral discourse suggests that self‐conscious emotions may be highly relevant to breastfeeding support interactions. However, the emotional impact of receiving support has not been fully explored. The aim of this review is to re‐examine qualitative UK research on receiving breastfeeding support, in order to explore the role of self‐conscious emotions and related appraisals in interactions with professional and peer supporters. From 2007 to 2020, 34 studies met criteria for inclusion. Using template analysis to identify findings relevant to self‐conscious emotions, we focused on shame, guilt, embarrassment, humiliation and pride. Because of cultural aversion to direct discussion of self‐conscious emotions, the template also identified thoughts about self‐evaluation, perceptions of judgement and sense of exposure. Self‐conscious emotions were explicitly mentioned in 25 papers, and related concerns were noted in all papers. Through thematic synthesis, three themes were identified, which suggested that (i) breastfeeding ‘support’ could present challenges to mothering identity and hence to emotional well‐being; (ii) many women managed interactions in order to avoid or minimise uncomfortable self‐conscious emotions; and (iii) those providing support for breastfeeding could facilitate women's emotion work by validating their mothering, or undermine this by invalidation, contributing to feelings of embarrassment, guilt or humiliation. Those supporting breastfeeding need good emotional ‘antennae’ if they are to ensure they also support transition to motherhood. This is the first study explicitly examining self‐conscious emotions in breastfeeding support, and further research is needed to explore the emotional nuances of women's interactions with supporters.

part in many women's infant feeding experiences and that understanding such feelings may be important for developing effective support (e.g., Benoit et al., 2016;Smyth, 2018;Thomson et al., 2015). This systematic review re-examines research on formal UK breastfeeding support to understand more fully the role of selfconscious emotions in women's interactions with breastfeeding supporters such as midwives, health visitors, lactation consultants, trained breastfeeding support volunteers and other mothers attending organised peer breastfeeding support schemes. Many motherinfant pairs struggle initially to establish breastfeeding and benefit from skilled support (McFadden et al., 2017). Within the United Kingdom, assistance with breastfeeding can be delivered via groups, individual consultation or telephone, in varying contexts including postnatal wards, women's homes, clinic settings and informal dropins offering specialised advice. Current guidelines on breastfeeding support recommend developing empathic relationships with mothers, which offer emotional support alongside information and technical advice on lactation and attachment at the breast (UNICEF UK, 2016a). However, despite these intentions, receiving support to overcome breastfeeding challenges may feel emotionally difficult to mothers who perceive the 'support' interaction as a situation that risks exposure of failure as a mother. Simply feeling 'not normal' as a woman or mother can trigger experiences of shame or embarrassment, particularly where there is a sense that others may discover this (Brown, 2007). Moreover, engagement with the gaze of health care workers can leave women feeling disapproved of and disempowered if they do not follow professionally approved feeding practices (Smyth, 2018). Therefore, interactions with those sanctioned as breastfeeding 'experts' may often be ripe for feelings of embarrassment, guilt, shame or even humiliation, rather than pride (Thomson et al., 2015).

| Self-conscious emotions
Although the range of self-conscious emotions and the distinctions between them are debated (Gibson, 2019), their importance is widely agreed. Guilt, pride, shame, humiliation and embarrassment (or their avoidance) motivate achievement, shape moral and caring behaviour and promote action to repair relationships and social image, though negative self-conscious emotions such as shame and guilt can be painful and debilitating (Tracy & Robins, 2007).
According to Gilbert (2017), shame is particularly problematic. It is the emotional resonance of believing ourselves flawed and devalued in the eyes of actual or imagined others, damaging our sense of connection and leaving us paralysed and isolated. He contrasts this with guilt-a sense that we have done something wrong. With its narrower focus on specific actions or omissions, guilt may not leave us feeling as powerless or damaged as shame, though it can be difficult to tolerate if we cannot make reparation, and both emotions have been linked with mental health difficulties such as depression (e.g., Kim et al., 2011). Both positive and negative self-conscious emotions share common features-a focus on the self as the object of either our own or others' evaluation and a concern with how our actions or selves align with standards, rules or identity goals that are often based on shared cultural understandings (Lewis, 2016;Tracy & Robins, 2007). Although it may not always be possible to draw sharp distinctions between experiences labelled as different self-conscious emotions (Gibson, 2019;Gilbert, 2003), these experiences can raise differing concerns.
For example, the term guilt is usually invoked when there is a breach of moral standards, whereas embarrassment occurs when publicly breaching less consequential social rules (Gibson, 2019) and generally involves a sense of exposure (Lewis, 2016). Robins and Schriber (2009) note that experiences referred to as shame or pride usually involve success or failure. For shame, the sense of a failed or flawed self may relate either to moral standards or to other domains, for example, the body, stigmatising attributes, performance failures that impact identity or being mocked (Leeming & Boyle, 2013). Loss of status or rejection is seen as particularly important by some theorists for understanding shame and humiliation and considered less important for guilt and embarrassment (e.g., Gilbert, 2003;Scheff, 2003). These varying emotional experiences prompt different actions ranging from appeasement, reparation and apology, to avoidance, defensiveness and boastfulness. However, a common feature of these actions is that they facilitate complex social goals relating to identity concerns and management of the self in social life (Tracy & Robins, 2007). Varieties of self-conscious emotion may therefore be expected in situations where we are aware of dominant cultural discourses about the kind of selves we ought to be and where measuring up to these moral standards or social conventions matters to us and, we assume, to those around us.

| Self-conscious emotions and motherhood
Motherhood-and more specifically infant feeding-might be one example. Mothering ideals may differ, but there is consensus that it is important to be a 'good mother' (Marshall et al., 2007). Women report that a sense of failure related to mothering is a common shame trigger (Brown, 2006), and several have argued that internalised, unrealistic expectations have made guilt an intrinsic part of motherhood (e.g., Liss et al., 2013). Contemporary notions of 'good mothering' can imply total focus on optimising the child's health, well-being and development, regardless of cost to the mother (Head, 2017), with the implication that mothers have a duty to breastfeed (Woollard & Porter, 2017). Infant feeding practices are therefore shaped not only by social conventions but also by moral injunctions (Russell et al., 2021). This creates a 'moral minefield' that mothers must negotiate to demonstrate to themselves and others that they are a 'good mother' regardless of feeding method Ryan et al., 2010). Although this opens the possibility for pride in achievement, infant feeding often takes place within social environments which are not supportive of breastfeeding, despite its widely recognised health benefits (see Victora et al., 2016, for review). Breastfeeding may not fit easily with other demands of family life  or local cultural contexts (Head, 2017). Women may feel embarrassed about public breastfeeding-as if they are disregarding the feelings of others (Boyer, 2018). On the other hand, some may feel shame or guilt about their perceived failure to establish breastfeeding (Williamson et al., 2012). These constraints on feeding choices are not always recognised and, instead, blame for using formula milk is internalised.
It is not surprising that, in relation to breastfeeding, women may feel 'shame if you do-shame if you don't' (Thomson et al., 2015, p. 33). However, sustained investigation of self-conscious emotions evoked by infant feeding is relatively recent, with just a few studies stating this as a primary aim, for example, exploring guilt in relation to public health messaging around breastfeeding (e.g., Benoit et al., 2016;Williams et al., 2012;Williams et al., 2013) or feeding method Komninou et al., 2017;Shepherd et al., 2017), or considering whether shame about perceived 'failures' with infant feeding may be mislabelled as guilt (e.g., Taylor & Wallace, 2012;Thomson et al., 2015) and a recent systematic review of research into shame and guilt in relation to feeding outcomes (Jackson et al., 2021). 1.3 | Self-conscious emotions in interaction with those supporting breastfeeding Self-conscious emotions often involve others (Gibson, 2019), so may be particularly likely when women discuss feeding their baby with those tasked to support their breastfeeding, potentially exposing perceived failures or successes and parts of the body deemed private. There is a sizeable body of research on women's experiences of receiving breastfeeding support, including several reviews (e.g., Joanna Briggs Institute, 2012;Schmied et al., 2011). However, to date, there has been limited empirical attention to self-conscious emotional processes within breastfeeding support, though studies contributing to these reviews have suggested that women can feel judged or objectified by those assisting with breastfeeding.
Despite the social value of self-conscious emotions, they can invoke tricky interpersonal dynamics (Sznycer, 2019). Therefore, understanding if and how self-conscious emotions shape support interactions might help improve the effectiveness of support. Displays of pride may be socially risky (Wubben et al., 2012), and guilt can lead to social avoidance (Yu et al., 2017). However, shame and humiliation may be the most problematic, sometimes resulting in blaming, hostility and anger (Elison et al., 2014;Elshout et al., 2017), or withdrawal from and avoidance of others (e.g., Chao et al., 2011), though we may have limited awareness of, or deny, these processes (Brown, 2006;Scheff, 2003). Therefore, self-conscious emotions are a slippery and taboo aspect of our lives to research (Scheff, 2016), which may influence relationships with others in ways that are problematic but unacknowledged. Lazare (1987) is one of the few to apply this to health care consultations. He argued that where clients or patients feel exposed, scrutinised or deficient, they may respond in ways that seem hostile or disrespectful, with the consequence of mutual blaming, labelling clients as 'difficult', and disengagement on both sides. This may be more likely where health care providers are sensitive to feeling shame themselves due to perceived criticism (Gilbert, 2017).
With a few notable exceptions (e.g., Dolezal, 2015;Gibson, 2015Gibson, , 2019Gilbert, 2017;Thomson et al., 2015), the management of self-conscious emotions in interactions with providers of health, social or maternity care has not been widely considered since Lazare's work. However, research outside the area of infant feeding suggests that others can play a role in helping individuals overcome problematic experiences of self-conscious emotion through interpersonal processes such as meaningful connection, empathy, acceptance and affirmation (e.g., Brown, 2006;Leeming & Boyle, 2013).

| Aim of paper
The aim of this paper is to re-examine qualitative research reporting UK women's perspectives on receiving formalised breastfeeding support to understand whether and how self-conscious emotions are relevant to interactions with those providing support. It is hoped that this will inform approaches to support that minimise difficult emotional experiences and complex interpersonal dynamics and support positive emotional responses to breastfeeding. As there is a recognised cultural aversion to direct discussion of self-conscious emotions (Scheff, 2016), and such feelings can be fleeting or avoided, we have used a broad conceptualisation of self-conscious emotions, including relevant appraisals. We focus on the ways in which women receiving breastfeeding support evaluate themselves and their actions, particularly where this appears to have emotional resonance. We also explore women's understanding of how those offering support have evaluated and treated them and their responses to this, including direct references to self-conscious emotions and data providing insight into how women managed emotion and protected their sense of self during these encounters.

| Approach taken to synthesis
This was a focused thematic synthesis, exploring aspects of previous research on women's experiences of breastfeeding support, using a new conceptual framework. As in 'framework-based synthesis' (Dixon-Woods, 2011), we used a conceptually derived set of a priori themes for initial organisation and synthesis of findings but followed the methods of King and Brooks' (2017) 'template analysis'. Therefore, unlike some forms of qualitative synthesis, we have not aimed to capture all aspects of receiving breastfeeding support interventions but to synthesise what this research reveals about self-conscious emotions.
Although variants of self-conscious emotions seem to be experienced universally, there are cultural differences in the issues about which people feel embarrassed, proud, guilty or ashamed, and the linguistic distinctions used to capture emotions (Gibson, 2019). We therefore limited the review to the United Kingdom, to enable better understanding of cultural context and of the support services researched.

| Identifying relevant papers
Primary research studies were identified that reported women's experiences of interactive postnatal support to breastfeed a healthy term baby, from either health care professionals or organised peer support (e.g., breastfeeding groups or trained peer supporters). Breastfeeding support could be the main focus of the study or part of a broader investigation, such as experiences of general postnatal support or experiences of breastfeeding. Asynchronous online support via social media was not included, though telephone support was.
The search was conducted systematically using five electronic databases: CINAHL, Scopus, MEDLINE, PsychINFO and PubMed.
Search terms included (Breastfeeding OR breast feeding OR breastfeeding OR lactation OR breastfeed*) AND (support OR postnatal support OR post-natal support OR lay support OR volunteer support OR social support OR breastfeeding counsellor OR breastfeeding education OR health education) AND (qualitative OR ethnography OR interviews OR phenomenology OR grounded theory OR thematic).
Limits were applied for English language and to retrieve peerreviewed publications. We searched for UK studies from January 2007, following a rigorously conducted meta-synthesis of women's experiences of breastfeeding support (Schmied et al., 2011), until 1 March 2020.
Two researchers independently screened titles, abstracts and fulltext articles for potential inclusion. Any discrepancies were discussed with a third reviewer. Studies reporting experiences of breastfeeding support for specific demographic groups such as young mothers and ethnic minority groups were included, but studies on groups or settings with high risk of infant feeding difficulties were not (e.g., neonatal intensive care, illicit drug use, low birthweight and specific health conditions). Studies were excluded if they only reported health professionals' perspectives or analysed data from a previously published study. Quality appraisal of each paper was carried out by one of the research team, and reviewed by another team member, using a checklist devised for the study. This enabled assessment of each paper against the following criteria for inclusion: appropriate and clearly reported recruitment and data collection procedures; ethical approval; sufficient in-depth qualitative data from mothers to provide insight into experiences of receiving breastfeeding support; and clear accounts of coding procedures, with sufficient quotes to justify and illustrate themes. As the aim was to reanalyse previous findings through a novel conceptual lens, studies were not rejected if the original analysis was mostly descriptive, with limited theoretical development, providing the data shed light on women's experiences of receiving support with breastfeeding and had been collected and reported in a rigorous and transparent manner. Due to the paucity of research on self-conscious emotions when receiving support, and the often hidden nature of these experiences, it was important not to exclude otherwise trustworthy studies unnecessarily. Figure 1 summarises the review process-34 papers were included in the synthesis (see Table 1).

| Data extraction
We were mindful of Scheff's (2016) argument that silence about shame and related emotions can extend to academic analyses, where the 's-word' may be ignored in favour of alternative interpretations of data. Therefore, our primary focus was original quotations and descriptive summaries of participants' accounts, though attention was also paid to authors' interpretations. For each paper, the findings section (or the parts of it related to support with infant feeding) was extracted for detailed coding in QSR's software package NVivo. We included findings on experiences related to support with formula feeding if support was from a service aimed primarily at encouraging breastfeeding, as is usually the case with infant feeding support within the United Kingdom. The remainder of the paper was examined (but not coded) to contextualise the data and findings.

| Method of analysis and framework for examining self-conscious emotions
Using template analysis (King & Brooks, 2017) enabled a focus on self-conscious emotions and related appraisals by including tentative a priori themes within a coding template that was revised and expanded in response to the data. Template analysis was originally developed for thematic analysis of primary data though there are some examples of its use for qualitative synthesis (e.g., Au, 2007;Brown & Lan, 2013). Analysis of most papers fitted with King and Brooks' (2017) description of using template analysis within a contextualist framework, given the focus on interpretation of experience and interactions in particular contexts. This aligns with an understanding of self-conscious emotions as interpretative and relational phenomena, comprising overlapping, loose clusters of embodied experiences with some similarities and also cultural variation (e.g., Gibson, 2019). Our use of terms such as 'shame', 'guilt' and 'humiliation' should therefore not be taken to imply that these are necessarily discrete, easily differentiated experiences. Drawing on the literature on self-conscious emotions, an initial outline coding template was developed, which comprised several overarching a priori themes (see Figure 2). These were 'soft' a priori themes (King & Brooks, 2017), being broad, loosely defined and intended only to draw attention to key aspects of the data-selfconscious emotions and self-evaluative processes-rather than necessarily shaping the final organisation of the data. Additional lower level themes and new overarching themes were then developed inductively through line-by-line coding of six papers, until a fuller template was agreed, which was then used for coding a further 18 papers. This is shown in Figure 3 and includes minor revisions to themes that were made collaboratively during coding of these 24 papers.
A more analytical and interpretative process of reviewing themes and data followed, where connections between themes and the context of studies were considered by the team. Some themes were decomposed, some relabelled and others reclustered, producing the final thematic structure (Figure 4), which was used for coding the remaining 10 papers. The themes were then audited and refined by one researcher rereading each of the 34 papers.

| FINDINGS
The relevance of self-conscious emotions to breastfeeding support was confirmed across the papers. Emotional aspects of receiving The review synthesised mothers' accounts of postnatal breastfeeding support. Several studies also included data from those providing support, other family members or observation (Condon & Salmon, 2015;Copeland et al., 2019;Edwards et al., 2018;Hoddinott, Craig, Britten, & McInnes, 2012;Hoddinott, Craig, MacLennan, et al., 2012;Hunt & Thomson, 2017;Ingram, 2013;Marshall et al., 2007;McFadden et al., 2013;Thomson & Crossland, 2019;. These data were not included in the synthesis nor were references to antenatal breastfeeding education or to postnatal support unrelated to infant feeding. support were more apparent in studies that focused primarily on women's experiences, rather than evaluation of an intervention or care context. However, even if briefly, all papers included material relevant to our broad conceptualisation of self-conscious emotions. Below, we begin with an overview of direct references to specific self-conscious emotions. This is followed by discussion of the final themes from the template analysis (precarious identity, managing the self and validation/invalidation). These themes were developed through exploration of issues relevant to self-conscious emotions, such as thoughts and feelings concerning self-evaluation, exposure in front of breastfeeding supporters, perceived judgement by others involved in feeding support and management of emotional feelings in relation to these issues. Cultural aversion to direct discussion of many self-conscious emotions means that these concerns are sometimes the only indicators of fleeting self-conscious emotions or avoidance of these. Thomson & Dykes, 2011). Although not often considered a specific self-conscious emotion, several papers also referred to feelings of self-consciousness in front of staff, visitors and family members, which seemed to have a similar meaning to embarrassment (Fox et al., 2015;Keely et al., 2015;Leeming et al., 2015;Ryan et al., 2017). References to pride were fewer, for example, when women felt they had succeeded in difficult circumstances (Brown & Lee, 2011;Condon & Salmon, 2015;Fox et al., 2015;Thomson & Dykes, 2011 3.3.1 | Cultural norms and expectations that clash with breastfeeding Encouragement to breastfeed could be counter to local cultural norms or social conventions (Condon et al., 2013;Entwistle et al., 2010;Fox et al., 2015;Marshall et al., 2007;Ryan et al., 2017;Tan et al., 2017).
A study of young mothers provided an example: 1 Some feared that to choose to breastfeed would transgress their own social norms: Midwives and doctors they try and persuade you to do it [breastfeed] because it is classed as a normal thing … but I don't know (Condon et al., 2013). particularly if they were overweight (Keely et al., 2015), inexperienced (Leeming et al., 2015;Ryan et al., 2017) or feeding an older child (Keevash et al., 2018).
Lack of familiarity could reduce confidence establishing breastfeeding, particularly for some groups, such as those on low incomes or younger mothers (Entwistle et al., 2010;Fox et al., 2015).
Breastfeeding could also feel morally dilemmatic if women felt it interfered with the expectations of a 'good' mother or partner, such as caring for other children (Hoddinott, Craig, Britten, & McInnes, 2012;Hunt & Thomson, 2017;Spencer et al., 2015), maintaining a sexual relationship (Marshall et al., 2007), allowing others to participate in bottle-feeding It took four or five weeks for her to regain her birth weight … and you kind of feel really guilty, don't you?
Like maybe I'm making the wrong choice here-maybe I'm being selfish by not giving her a bottle (Fox et al., 2015).
At times, these concerns, which were at odds with the advice from infant feeding advisors, were prompted by comments from sceptical family members and friends:

| Additional factors undermining confidence in mothering
Women's confidence as mothers, and therefore presumably their interactions with health care providers, was also undermined by  (Keely et al., 2015). There was a recurrent theme of women lacking confidence in their ability to breastfeed (see particularly Entwistle et al., 2010;Leeming et al., 2015;Marshall et al., 2007). 3.3.4 | Concluding comments on precarious identity as a breastfeeding mother Infant feeding could present considerable challenges to the task of becoming a 'good' mother, creating guilt, embarrassment or a sense of shameful inadequacy. However, several studies also reported breastfeeding contributing to feelings of bonding, pride and achievement, even where there were initial difficulties (Brown & Lee, 2011;Entwistle et al., 2010;Fox et al., 2015;Guyer et al., 2012;Jardine et al., 2017;Lagan et al., 2014;Leeming et al., 2015;Marshall et al., 2007;Thomson & Crossland, 2013;Thomson & Dykes, 2011 However, there was some ambivalence as pride could be seen as 'silly' (Fox et al., 2015) or inappropriate because breastfeeding was 'normal' (Brown & Lee, 2011).
This theme suggests that although breastfeeding can be an important part of developing a positive mothering identity, this can be precarious. For many women, it was not always clear how to be a 'good' mother with regard to feeding their baby, because breastfeeding challenged their moral and cultural context. The additional challenges of new parenthood, including childbirth itself, further undermined their ability to navigate these contradictory pressures and could place them in an emotionally precarious position when engaging with breastfeeding support.

| Theme 2: Managing the self in interaction with breastfeeding supporters
These conclusions do not necessarily mean that women across the studies experienced strong negative self-conscious emotions, though some did. Many enjoyed breastfeeding and felt well supported.
Several studies also found women actively engaging in what has been called 'emotion work' (Hochschild, 1979) or 'moral work' (Ryan et al., 2010)  feeding (e.g., Condon et al., 2013;Hoddinott, Craig, Britten, & McInnes, 2012;Hunt & Thomson, 2017;Marshall et al., 2007;Spencer et al., 2015). However, these modifications were not compatible with exclusive breastfeeding. Sometimes infant feeding advisors supported these attempts to balance different pressures, but where women thought that supporters might disagree, they sometimes withdrew or avoided discussion (Hoddinott, Craig, Britten, & McInnes, 2012;Spencer et al., 2015;Taylor et al., 2019). Some mothers spoke of reluctance to engage with additional support, such as peer support, for fear they would not 'measure up' (Fraser et al., 2020;Hunt & Thomson, 2017;Keely et al., 2015;Ryan et al., 2017;Spencer et al., 2015;Taylor et al., 2019;Thomson & Dykes, 2011). Hunt and Thomson (2017) commented on this avoidance of potentially shaming exposure: The conflict between women's self-perceptions of being 'a failure' doing 'a crappy job' and the imagined 'successful' breast-feeders who accessed the [peer-support] groups was a key barrier to access.
However, for other women, contact with other breastfeeding women could be a strategy for strengthening identity and confidence as breastfeeding mothers (see Section 3.5.1). This could involve shifting alliances as they avoided previous friends and made new friends, for example, via Baby Cafés: I thought I need to find a place to go to where I've got likeminded people and that grew my confidence to be honest as a new mum and as a breastfeeding mum (Fox et al., 2015).

| Being a 'good mother-client'
Concerns about impression management included behaving appropriately as a client of maternity services. Some mothers 'felt guilty about bothering staff' (Beake et al., 2010), particularly on busy postnatal wards (Beake et al., 2010;Hoddinott, Craig, Britten, & McInnes, 2012;Hunter et al., 2015;Redshaw & Henderson, 2012;Ryan et al., 2017) and also in community services (Hoddinott, Craig, MacLennan, et al., 2012;Thomson & Crossland, 2013). They avoided pressing the buzzer, or asking for help, to avoid the guilt or embarrassment of becoming a 'nuisance'. However, this reduced the likelihood of effective support. In one study, a younger mother explained the creative strategies she had employed: Sometimes I used to press the buzzer, sort of put it back, and they'd come and I'd say 'oh, oh, I must have leant on it!' And then I'd say 'oh, while you're here … ', because I just felt like I was being such a nuisance (Hunter et al., 2015). A stated intention to 'try' breastfeeding was noted by Condon et al. (2013) and Spencer et al. (2015) as another way of being a good mother-client, enabling women to claim good intentions, while leaving an exit route open, 'rather than make a choice to formula feed from the start' (Spencer et al., 2015).
Some strategies for managing emotion by being a good client were more straightforward and involved seeking 'approval' for infant feeding decisions directly from health care professionals or avoiding actions they assumed were not sanctioned (Entwistle et al., 2010;Guyer et al., 2012;Hunter et al., 2015;Lagan et al., 2014;Leeming et al., 2015;Spencer et al., 2015).

| Interpretation and reinterpretation
Emotion work also included reinterpreting events, which challenged maternal identity. As discussed under Section 3.5.1, breastfeeding supporters sometimes facilitated this, though positive interpretation could be an internally driven process, focusing on achievements or adopting a 'can do' approach (Brown & Lee, 2011;Entwistle et al., 2010;Jardine et al., 2017;Keely et al., 2015;Ryan et al., 2017): Some justifications for discontinuing breastfeeding also preserved self-esteem by attributing this to lack of support from health care professionals and limited or misleading information (Beake et al., 2010;Fox et al., 2015;Keely et al., 2015;Redshaw & Henderson, 2012). Whether or not these accounts are 'true', they indicate the pressure felt by women to account for their infant feeding decisions and direct blame somewhere. Hoddinott, Craig, Britten, and McInnes (2012) noted that some families expressed doubts about the improved health outcomes of breast milk compared with formula milk, concluding that this reinterpretation served to 'to counter any implications of being a 'bad mother' or putting a child at risk'. Criticisms of health care advice and staff were only expressed as overt anger in one of the two diary studies (Taylor et al., 2019). Nonetheless, for a few women, this interpretative emotion work led to entrenched positions where breastfeeding became 'unrealistic', formula feeding became 'reasonable' and those advocating breastfeeding were to be resisted, avoided or blamed.

| Concluding comments on managing the self
Engagement with breastfeeding support was not a straightforward process for many women as it could mean far more to them than simply receiving infant feeding advice and guidance. For many, their mothering identity was at stake and interactions needed careful management to maintain emotional well-being.

| Theme 3: Validation/invalidation
Interactions with peer supporters and health care professionals could support emotion work by positively affirming women's mothering and countering self-doubt, guilt and shame. However, many women indicated that care environments had exacerbated negative feelings.
Below, we discuss this by considering whether women felt validated or invalidated by those offering support and guidance. The term 'validation' does not necessarily mean that a woman felt supporters agreed with her infant feeding decisions but that she thought they respected and accepted her and perceived her decisions to be meaningful and understandable given her circumstances. We use the term 'invalidation' to capture not just an absence of validation but a sense from women's accounts that others devalued and disrespected them, were disinterested or critical, and undermined their identity as a mother.

| Validation
Most papers reported some participants saying breastfeeding support had helped them to feel better about their mothering, through two overlapping processes of validation: (i) reflecting back a positive view of the woman and her mothering and (ii) meaningful connection and acceptance.

Positive feedback
This first form of validation could promote feelings of capability, achievement or pride and help to counter potential shame, guilt or embarrassment. Postive feedback could be simple reassurance that feeding was successful and the baby gaining weight (e.g., Taylor  Affirmation of women's ability to breastfeed could validate them more generally as mothers: … they [peer supporters] tell me, you're doing a great job and it just makes you feel great. Oh I'm being a good mum, I'm being a great mum, I'm doing it right, it's nice (Thomson, Crossland, & Dykes, 2012).
Health care professionals could be particularly helpful in countering negative self-belief. They could offer reassurance that breastfeeding difficulties were not unusual, were surmountable, could be temporary and were not indicative of deficient or incompetent mothering (Condon et al., 2013;Hinsliff-Smith et al., 2014;Hoddinott, Craig, Britten, & McInnes, 2012;Hoddinott, Craig, MacLennan, et al., 2012;Ingram, 2013;Leeming et al., 2015;Marshall et al., 2007;Tan et al., 2017;Taylor et al., 2019). Peer supporters could also contextualise, reinterpret and normalise difficulties, countering a sense of failure: She made me feel that I was not the only person in the world having this problem, I failed to feed my daughter and … thought breastfeeding was natural, I thought it must be me. So having someone explain it is normal and not completely easy for everyone (Thomson & Crossland, 2013; see also Brown & Lee, 2011;Copeland et al., 2019;Fox et al., 2015;Tan et al., 2017;Thomson & Crossland, 2019;Thomson, Crossland, & Dykes, 2012).
Peer support groups could play an additional role in reflecting a positive view of breastfeeding, by constructing an environment where 'breastfeeding was the norm' (Tan et al., 2017;see also Copeland et al., 2019) and where breastfeeding was valued and shared, including the challenges (see also Brown & Lee, 2011;Fox et al., 2015;Fraser et al., 2020;Guyer et al., 2012;Thomson, Crossland, & Dykes, 2012;Thomson & Dykes, 2011). Therefore, although peer contact required women to risk feeling that they did not measure up (see Section 3.4.1), it could be an important means of normalising and validating breastfeeding and resisting feelings of deviance or embarrassment.

Connection and acceptance
Positive experiences of peer support also supported women's emotion work by providing validation through connection with and acceptance by others-the opposite experience to shame. Most studies noted the importance of peer or professional supporters having a warm, patient, kind, respectful or caring manner. Some studies highlighted what Hoddinott, Craig, Britten, and McInnes (2012) call a 'woman-centred' approach, which does not focus solely on maintaining breastfeeding but also shows that women are valued, by considering their individual situations and needs. Women appreciated breastfeeding supporters trying to understand their perspectives, concerns and experiences in a non-judgemental manner, which respected their choices (Entwistle et al., 2010;Fox et al., 2015;Ingram, 2013;Tan et al., 2017;Taylor et al., 2019;Thomson & Crossland, 2013 and promoted agency (Ryan et al., 2017). This not only helped to bolster positive self-appraisals and related emotional experiences but could also enable disclosure of concerns about taboo topics (Ingram, 2013;Tan et al., 2017;, which allows supporters to challenge any negative self-beliefs-the core of negative self-conscious emotions. Where supporters accepted feeding practices that did not meet breastfeeding ideals but enabled some breast milk to be offered, this could help counter a sense of guilt or failure and acknowledge a woman's efforts in challenging circumstances (Hoddinott, Craig, Britten, & McInnes, 2012;Thomson & Crossland, 2013;Thomson, Crossland, & Dykes, 2012;Wade et al., 2009). This woman-centred approach left mothers feeling that they and their wider concerns mattered, rather than just their ability to feed their baby optimally, though Entwistle et al. (2010) note the potential of this approach to undermine breastfeeding. Women valued other behaviours that indicated they mattered to supporters, such as spending time with them, being readily available, initiating contact and actively listening (Beake et al., 2010;Copeland et al., 2019;Hoddinott, Craig, Britten, & McInnes, 2012;Hunter et al., 2015;Ingram, 2013;Ryan et al., 2017;Tan et al., 2017;Thomson & Crossland, 2013Thomson, Crossland, & Dykes, 2012;.  found that women receiving individually tailored gifts as incentives for breastfeeding interpreted these as a sign that they mattered as a person rather than a feeding agent. The gifts were 'connectors', communicating that the peer support scheme 'cared about' the women.

| Invalidation
Despite the importance of validation, many papers reported accounts from at least one or two women of interactions that could lead to shame, embarrassment or humiliation through women being demeaned or disregarded by those who were meant to support their breastfeeding. Redshaw and Henderson (2012) summarised several participants' portrayal of negative staff attitudes when supporting infant feeding on postnatal wards: Negative staff attitudes or behavior ranged from women feeling 'bullied' or 'judged', being 'shouted at', midwives acting in a 'condescending manner', being 'insensitive', 'inconsiderate', 'disrespectful', and 'rude'.
Women also complained of being disrespected as autonomous adults through pressure to make infant feeding decisions, dismissal of their ideas, judgemental surveillance or infantilisation (Condon et al., 2013;Fox et al., 2015;Guyer et al., 2012;Hinsliff-Smith et al., 2014;Hoddinott, Craig, Britten, & McInnes, 2012;Hunt & Thomson, 2017;Hunter et al., 2015;Keevash et al., 2018;Lagan et al., 2014;Leeming et al., 2015;Marshall et al., 2007;Redshaw & Henderson, 2012;Ryan et al., 2017;Spencer et al., 2015;Taylor et al., 2019;Thomson & Crossland, 2013Thomson & Dykes, 2011). Such experiences could 'make you doubt yourself' (Thomson & Crossland, 2019) and therefore might be ripe for feelings of embarrassment, guilt, shame or humiliation, though emotional responses were often implied or left unstated in interview data. Drawing on detailed video diaries, rather than the interviews used in most studies, Taylor et al. (2019) noted 'deep emotional turmoil' and anger in response to feeling 'scrutinised and humiliated' by staff and suggested that the extent of these feelings would have been difficult to capture in an interview.
Some women described more subtle forms of invalidation where they felt uncared for because staff seemed disinterested in their needs or unavailable (Fraser et al., 2020;Ryan et al., 2017;Taylor et al., 2019) or they felt disrespected because their needs and preferences were not acknowledged, for example, with regard to breastfeeding support, privacy or rest (Beake et al., 2010;Hunter et al., 2015;Keely et al., 2015;Redshaw & Henderson, 2012;Ryan et al., 2017). Although these various forms of invalidation were experienced within community and peer support services, they were more frequently discussed in relation to hospital care. For example, Hunter et al. (2015) referred to: Task-based, routinised care, during which the young women were treated as objects rather than selfdetermining individuals … (see also Taylor et al., 2019).

| Concluding comments on validationinvalidation
The first two themes illustrated how some women approached motherhood and infant feeding feeling under surveillance and expecting to be judged by maternity care workers, particularly if not breastfeeding. Taylor et al. (2019) linked this to a medicalised, risk averse and authoritarian approach to infant feeding-checking if mothers and babies were on the 'right track'. Thus, prior expectations of scrutiny may have amplified any subtle signals of disrespect, pressure or blame from staff, and as Redshaw and Henderson (2012) note, 'a single unpleasant incident sometimes marred postnatal care'. As identity was already precarious for some, and negative self-appraisals evident, they did not need to experience active invalidation in order to struggle with an absence of validation and reassurance, and the emotional consequences of this.

| DISCUSSION
This was a focused synthesis, exploring self-conscious emotions, and was not an attempt to capture all aspects of women's experiences of receiving breastfeeding support discussed in the original papers.
However, the review clearly suggests that self-conscious emotions and related self-appraisals, particularly negative ones, are relevant to many UK women's experiences of varied types of breastfeeding support. Though women may not always discuss these emotions explicitly in interviews, or experience them strongly, many women can find their mothering identity precarious and in need of active management when interacting with supporters. Careful emotion work is required to avoid feeling guilt, shame, embarrassment or even humiliation. As indicated in a recent review (Russell et al., 2021) Thomson et al.'s (2015) conclusion that such cultures can simultaneously shame mothers for breastfeeding and shame them for formula feeding or struggling with breastfeeding. Although there were no explicit references to shame in the reviewed papers, there were many references to feeling inadequate, deviant, exposed, excluded or a failure. This supports the contention that discussion of guilt or discomfort around infant feeding may sometimes mask unspoken experiences of shame (Leeming, 2018;Taylor & Wallace, 2012 (Trickey, 2016). Challenging the idea of fixed all-or-nothing breastfeeding 'rules', which create a perception of failure when these are broken, may also be helpful. Likewise, presenting a more nuanced understanding of breastfeeding as a 'natural' process may prevent breastfeeding being idealised and romanticised in a way that erases the possibility of difficulties and means women internalise challenges as personal deficiencies (Leeming, 2018). Instead, it seems useful to communicate expectation of an 'adjustment period' (Trickey & Newburn, 2014) to women and their families. The theme 'precarious identity' also points to the importance of working with families or community groups to shift devaluation of breastfeeding and challenge the idea that it might be self-indulgent or exhibitionist. Some women seemed to be turning to formula milk to be 'good' mothers within their social networks, rather than because of an initial preference for this. However, the promotion of pride in breastfeeding may need careful thought. An overemphasis on pride in personal achievement raises the possibility of shame where there is perceived failure (Shepherd et al., 2017). Nonetheless, pride may be useful for asserting a collective identity as breastfeeding women (Mecinska, 2018), and it is worth unpacking the different meanings of 'pride', as some may be more helpful than others (Wubben et al., 2012).
There is increasing recognition of the importance of respectful conversations with mothers about infant feeding, which are empathic, person-centred and non-judgemental (e.g., Schmied et al., 2011;UNICEF UK, 2016a). This review extends understanding of why this validating approach is useful and fits with Gilbert's (2017) argument that compassion, kindness and acceptance can counter feelings of shame, embarrassment and humiliation. Validation not only boosts confidence but also reduces the risk of emotional responses that disrupt relationships between mothers and supporters. Recently developed peer support interventions are promising in that they explicitly incorporate a validating woman-centred ethos and consideration of emotional responses, based on the Solihull Approach (Tan et al., 2017), motivational interviewing (Copeland et al., 2019) or an assets-based approach (Ingram et al., 2020, published after present review). However, our review suggests that some environments, particularly inpatient postnatal settings, may facilitate invalidation rather than validation. Those supporting infant feeding need good emotional 'antennae' because, although self-conscious emotions are common, they can go 'underground', particularly shame, and wreak unacknowledged havoc (Scheff, 2003), through the disengagement, subterfuge or hostility discussed above. Self-conscious emotions are not clearly expressed via facial expressions (Robins & Schriber, 2009), and Gibson's (2015) analyses of social work suggest that shame and related feelings may be concealed in defensiveness, withdrawal, indirectness, hesitation and negative self-evaluation. Those offering support may need to be alert to these cues. Although women sometimes referred to breastfeeding supporters as good listeners, none of the studies reviewed explicitly mentioned participants discussing selfconscious emotions with supporters other than family/friends. However, this may have been unreported. Research on resilience to shame, probably the most toxic of the self-conscious emotions, suggests that articulating shame may be key to tolerating it, creating opportunities for others to offer affirmation and alternative interpretations (Brown, 2006). Good emotional 'antennae' might prompt discussion of emotions and also guide maternity care workers' decisions about whether women are likely to engage comfortably with more exposed settings such as peer support groups.

| Future research
Further research into the emotional impact of breastfeeding support interventions might help to shed light on the question raised previously as to why particular breastfeeding support interventions may be successful sometimes, but not others (e.g., Hoddinott et al., 2011).
The current findings illustrate how interventions or support services that appear similar in design may have very different emotional effects for service users, depending on the relationships between those giving and receiving support (Leeming et al., 2017). It may be the case that researching how breastfeeding support is experienced by different women within different relationships and contexts is as important as researching what works, when 'what' is defined in broad brush terms as a packaged intervention. In-depth ethnographic examination of how the cultural or organisational context shapes support relationships and behaviours might also enable better understanding of how some postnatal inpatient settings foster invalidating interactions whereas others foster validation. Gibson (2019) notes how organisational practices shape not just relationships but possibilities for shame and pride for the professionals working within them and their clients and how the experience of each party influences that of the other. The sense of guilt, failure, shame or pride that those supporting breastfeeding may feel in relation to organisational infant feeding targets has been outside the scope of this review. However, further exploration of the co-construction of self-conscious emotions between mothers, those supporting them and the organisations they work within may help to minimise devaluing interactions and ensure that breastfeeding interventions truly support breastfeeding mothers.

| Evaluation of study
By beginning with a clear conceptual framework for our synthesis, we offer a new lens on breastfeeding interventions. This does mean that our approach draws attention to some issues rather than others. However, these issues may often be hidden and underresearched due to their taboo nature. This is the first review to explore the role of self-conscious emotions in breastfeeding support. However, it should be seen as opening an area for further research rather than offering a detailed examination of emotion processes. We used a broad conceptualisation of selfconscious emotions, focusing on accounts of self-evaluation and concern about one's worth in the eyes of others, whether or not emotions were mentioned explicitly. Our aim was to include experiences where selfconscious emotions might be on the periphery of consciousness but avoided through careful self-management. This approach was important given the cultural preference for indirect references to some selfconscious emotions, the assumption that unacknowledged selfconscious emotions can still profoundly shape interactions (Scheff, 2003) and the fact that many of the papers reviewed had not intended to probe emotion processes in depth. However, this inclusive approach makes it difficult to distinguish between self-evaluations that were accompanied by noticeable affect and those that were not. Further research, using methods that enable more in-depth recounting of troubling and taboo emotions, will help us to understand more fully how self-conscious emotions are felt by women in the context of receiving breastfeeding support, how this can shape their interactions with health care providers and how these interactions can be rethought to ensure they support mothers and protect breastfeeding.

ACKNOWLEDGMENT
The study was unfunded but supported by scholarly time provided to the authors by their employer-University of Huddersfield.