Social Neuroscience - Part 3

Relationships

© 2018 The Neuroscience School

Human relationships primarily consist of friends, family, and a romantic partner. We invest a huge amount of time and effort into cultivating and maintaining these relationships.

Why? Because we like it, and we like it because it is good for us, not only, as we saw in part 2, for the material benefits that accrue from cooperation, but also because it has protective effects on our health. Social support in the form of supportive family interactions and the presence of an intimate and confiding relationship has a protective effect against cardiovascular-, immune-, and endocrine-related health issues. In contrast, loneliness and lack of intimacy may have the opposite effect: for instance, being associated with greater cognitive decline in old age.

In Part 3 here, we’ll look at love: whether it is a unitary concept and how it may be represented in the brain. We’ll also take a look at attachment, focusing on the infant-parent bonds and romantic bonds. And finally, we’ll look at separation, social exclusion, and loneliness.

Uchino, BN, & bulletin, C.-J. (1996). The relationship between social support and physiological processes: a review with emphasis on underlying mechanisms and implications for health. Psychological bulletin. Retrieved from http://psycnet.apa.org/journals/bul/119/3/488/

Wilson, R. S., Krueger, K. R., Arnold, S. E., Schneider, J. A., Kelly, J. F., Barnes, L. L., Tang, Y., et al. (2007). Loneliness and risk of Alzheimer disease. Archives of general psychiatry, 64(2), 234–40.

The most influential theory of love is Sternberg’s triangular theory of love. In this model, different types of love arise from three different factors: passion, intimacy, and commitment. They combine in different ways to generate different kinds of love. So, for example, passion and intimacy combine to create romantic love; intimacy and commitment combine to create companionate love; and fatuous love is a combination of passion and commitment.

Of course, the type of love experienced may change over time. The passionate component of love, for example, seems to last only six months to three years of a relationship. Many relationships continue to endure for decades in which other aspects of love, such as intimacy and commitment, may grow.

By en:User:Lnesaderivative work: Rafy (Triangular_Theory_of_Love.gif) [Public domain , via Wikimedia Commons

Proximity

Familiarity

Physical attractiveness

Similarity of attitudes

Reciprocal liking

What factors determine whom we fall in love with?

The rules of attraction have been extensively studied by psychologists. And many of these factors apply to friendships as well as romantic relationships.

1. Proximity. People are more likely to form friends with people on the same floor relative to other floors or other buildings.

2. Familiarity. People are more liked when they are seen more often.3. Physical attractiveness. The matching hypothesis states that people are more likely to form long-standing

relationships with those who are as equally physically attractive as they are.4. Similarity of attitudes. In a study by Newcomb (1961) where students noted down the people they were attracted

to, initially attraction was related to proximity but then shifted to similarity of attitudes.5. Reciprocal liking. If you are told that someone likes you then you are more likely to like them.

Newcomb, F. M. (1961). The Acquaintance Process. New York: Holt, Rinehart & Winston.

Love is an emotion that is elicited in the presence of or the thought of another person with whom there is an attachment. There appear to be different facets to love and there is evidence for a common circuit for different kinds of love. A mother’s love for her child, for example, is associated with increased activity in a number of regions rich in oxytocin and vasopressin, and in reward centres such as the striatum. However, when viewing images of one’s own child compared to a familiar child acquaintance, there were deactivations in the amygdala and regions that have traditionally been linked with mentalizing, such as temporal poles, temporo-parietal junction, and medial prefrontal cortex. These are also the same areas of deactivation found in a previous study that the authors did on romantic love. The authors argue that love involves a pull-push mechanism in the brain by which regions involved in critical social assessment are deactivated while regions involved in reward and attachment are activated. This may provide the basis for the unconditional nature of love that distinguishes it from emotional responses that are context sensitive.

Bartels, A., & Zeki, S. (2004). The neural correlates of maternal and romantic love. Neuroimage, 21, 1155-66.

As well as being rewarding, the mere presence of a loved one may also act as a buffer against stress or pain. Here’s an example: When women were shown pictures of a male partner while they were receiving a painful (but non-harmful) stimulus, their reported intensity of the painful stimulus was reduced. You can see that here, compared to viewing photos of strangers or objects. And this analgesic effect was correlated with activity in a number of regions, including the reward-related nucleus accumbens.

Master, S. L., Eisenberger, N., I., Taylor, S. E., Naliboff, B. D., Shirinyan, D., & Lieberman, M. D. (2009). A picture’s worth: partner photographs reduce experimentally induced pain. Psychological Science, 20, 1316-18.

Younger, J., Aron, A., Parke, S., Chatterjee, N., & Mackey, S. (2010). viewing pictures of a romantic partner reduces experimental pain: Involvement of neural reward systems. Plus One, 5, 7.

0

1

2

3

4

5

6

7

8

Males Females

In love Control

The term “falling in love” implies a lack of control and some researchers have compared the feeling to the manic phase of bipolar depression, and the craving and intrusive thoughts about the loved one to obsessive-compulsive disorder (OCD). Of course, this doesn’t necessarily reflect mental disorders but, instead, may be an important biological mechanism that enables us to put aside inhibitions and abandon our comfort zones, such as neglecting family and friends to spend time with a new partner.

There are also hormonal changes in the early passionate phase of a relationship rather than the later companionate stages. Testosterone levels are lower in males and higher in females during early stages of a romantic relationship, but return to normal 12-18 months later, even when they stay in the same relationship.

Marazziti, D., & Canale, D. (2004). Hormonal changes when falling in love. Psychoneuroendocrinology, 29, 931-936.

And oxytocin levels are high in the initial stages or romantic love and correlate both with signs of affection and also relationship-related worries. And oxytocin levels at the beginning of the study predicted whether the couples would stay together 6 months later, or separate: those with higher oxytocin levels at the beginning of the study stayed together compared to those who had lower oxytocin levels.

Oxytocin is important for forming relationships. However, there is some doubt about whether this hormone specifically services this function, or whether it has a wider and more general role in social cognition such as anxiety reduction. Research has shown, for example, that oxytocin, both naturally occurring and externally administered has a wider function: It reduces anxiety pointing to the importance of close relationships for mental health.

Schneiderman, I., Zagoory-Sharon, O., Leckman, J. F., & Feldman, R. (2012). Oxytocin during the initial stages of romantic attachment: Relations to couples’ interactive reciprocity. Psychoneuroendocrinology, 37, 1277-85.

Let’s move on to one of the most important, if not the most important, determinant of the quality of adult relationships: the quality of early infant and early childhood attachment to a caregiver. Attachment is a long-enduring, emotionally meaningful tie to a particular individual. Attachments in young children provide them with comfort and security and is found in all animal species in which the young are initially in need of care and protection. This is vividly illustrated by the case of zoologist Kondrad Lorenz and his chicks. Newly hatched chicks imprint on the first moving object they see. Here, the chicks have imprinted on Konrad Lorenz and follow him around. Once imprinted, the bird is virtually unable to learn to follow a new foster parent. Here’s a short clip about Konrad Lorenz and how he discovered imprinting in birds.

Individual At onset of attachment At 18 months

Mother 95 81

Father 30 75Grandparent 11 45

Other relative 8 44

Sibling 2 24

Other Child 3 14

Percentage of Infants Forming Attachments to Particular Individuals

In infants, who they attach to is rather more flexible. The mother, though, is the first and strongest attachment relationship which emerges at around 7-8 months of age. At this age, infants tend to experience a fear of strangers, and this is true across a wide range of cultures. But by 18 months, they’ve widened their attachment relationships. At this age, most infants have developed an attachment to their father. You can see on the chart here, that many are also able to form attachments to grandparents and other relatives, and to siblings.

Schaffer, H. R., & Emerson, P. E. (1964). The development of social attachments in infancy. Monographs of the Society for Research in Child Development, 29, 1-77.

Some the earliest studies on attachment were carried out by psychologist Harry Harlow in the 1950s. He looked at the effect of maternal deprivation in rhesus monkeys. These were landmark studies, not only in primatology, but in the then evolving science of attachment and loss. These monkey-love experiments had powerful implications for separations of mothers and infants, including adoption, and childrearing in general.

Harlow showed that mother-love was emotional rather than physiological. Second, he showed that capacity for attachment was closely associated with critical periods in early life, after which it was difficult or impossible to compensate for the loss of the initial emotional security. The critical period thesis confirmed the wisdom of placing infants with adoptive parents as shortly after birth as possible. Harlow’s work provided experimental evidence for prioritizing psychological over biological parenthood.

How did Harlow go about constructing his science of love? He separated infant monkeys from their mothers a few hours after birth, then arranged for the young animals to be “raised” by two kinds of surrogate monkey mother machines, both equipped to dispense milk. One mother was made out of bare wire mesh. The other was a wire mother covered with soft terry cloth. Harlow’s first observation was that monkeys who had a choice of mothers spent far more time clinging to the terry cloth mothers, even when their milk came from bottles from the bare wire mothers. This suggested that infant love wasn't a simple response to the satisfaction of physiological needs. Attachment was not primarily about hunger or thirst. It could not be reduced to nursing.

Then Harlow modified his experiment and made a second important observation. When he separated the infants into two groups and gave them no choice between the two types of mothers, all the monkeys drank equal amounts and grew physically at the same rate. But the similarities ended there. Monkeys who had soft, tactile contact with their terry cloth mothers behaved quite differently than monkeys whose mothers were made out of cold, hard wire. Harlow hypothesized that members of the first group, those with the terry cloth mothers, benefitted from a psychological resource—emotional attachment—that wasn’t available to members of the second group, the bare wire mothers. This showed that cuddling provided reassurance and security to infants and kept normal development on track.

There is some evidence that attachment styles are partially conserved from infancy into adulthood. A 22-year long study examined the effects of infant attachment style on adult emotional regulation. In this study, participants were shown emotionally positive and negative pictures while their brains were being scanned and they judged the image as positive or negative using a button press. They were then instructed to either ‘increase’ or ‘decrease’ their emotional response to the picture, or to continue to simply ‘attend’ to the image. The researchers found that, compared to just attending, the insecure attachment group showed greater activation in the rostral anterior cingulate cortex, dorsal medial prefrontal cortex and the right and left anterior prefrontal cortex when trying to up-regulate a positive emotion. The rostral ACC and dm PFC are components of the emotion regulation circuitry that have influence over the nucleus accumbens (we’ve already seen that the nucleus accumbent is associated with reward). And indeed this was found for the secure attachment group: there was a positive association between the rACC, the dmPFC and the nucleus accumbens activation that was stronger for the secure attachment group than the insecure attachment group.

In short, the brain imaging found that participants who previously had insecure attachments as infants required more prefrontal activation (lateral and medial prefrontal cortex and anterior cingulate) to reappraise positive images. This points to a relative inefficiency in the brain regulation of positive affect. Those with previously secure attachments showed greater connectivity between these regions and the reward-related nucleus accumbens.

This study shows that the brain’s circuitry involved in emotion regulation may be directly shaped by the quality of the early environment and this is shaped by the quality of infant attachment. Disturbances in the mother–infant relationship may persistently alter the brain circuitry of emotion regulation, and may affect emotional adjustment in adulthood.

Moutsiana, C., Fearon, P., Murray, L., Cooper, P., Goodyer, I., Johnstone, T., & Halligan, S. (2014). Making an effort to feel positive: insecure attachment in infancy predicts the neural underpinnings of emotion regulation in adulthood. Journal of Child Psychology and Psychiatry, 55(9), 999–1008.

Insecure AttachmentAvoidant:

"I am comfortable without close emotional relationships”

"I am somewhat uncomfortable getting close to others. I want emotionally close relationships, but I find it difficult to trust others completely, or to depend on them. I sometimes worry that I will be hurt if I allow myself to become too close to other people."

Anxious:

"I want to be completely emotionally intimate with others, but I often find that others are reluctant to get as close as I would like"

Under normal circumstances, social interactions with beloved ones (children, parents, partners), friends, or any “significant” (e.g., contextually relevant) other person with a cooperative relationship (e.g., joint task), all enhance the experience of positive emotions and increased activity in the brain’s reward circuits. This could contribute to promoting expectations of positive social outcomes, and thus in turn enhance attachment behaviours and a feeling of safety.

Insecurely attached individuals show a different pattern. But before we go into how they differ, let’s first look at the two basic types of insecure attachment: avoidant and anxious. Avoidant individuals tend to be ambivalent towards relationships. They would agree with statements such as "I am comfortable without close emotional relationships” or "I am somewhat uncomfortable getting close to others. I want emotionally close relationships, but I find it difficult to trust others completely, or to depend on them. I sometimes worry that I will be hurt if I allow myself to become too close to other people.” They view themselves as self-sufficient and invulnerable to feelings associated with being closely attached to others. They often deny needing close relationships. They tend to feel uncomfortable with emotional closeness. These mixed feelings are combined with sometimes unconscious, negative views about themselves and their attachments. They rate positive social information as less arousing and less pleasant, compared to securely attached individuals. This seems to suggest that avoidant individuals purposely keep their attachment system in a dulled state.

Anxious individuals, on the other hand, would agree with statements such as “I want to be completely emotionally intimate with others, but I often find that others are reluctant to get as close as I would like.” People with this style of attachment seek high levels of intimacy, approval, and responsiveness from their attachment figure. They sometimes value intimacy to such an extent that they become overly dependent on the attachment figure. Research has found that they have increased activation of the aversion system in response to negative social cues. This reflects the tendency of anxiously attached individuals to experience increased distress in situations of personal failure or social disapproval, when social support would be desired instead.

In securely attached individuals, on the other hand, (mutual) social interactions seem generally to be associated with

Ventral striatum left

Avoidance scoresAmygdala left

Anxious scores

Ventral striatum left

Amygdala left

Let’s look at the consequences of attachment style on social interaction. Here’s a study that looked at the influence of attachment styles on the processing of negative and positive social stimuli in attachment-related contexts.

Participants looked at dots on a screen while their brains were being scanned. The screen was divided in half and the number of dots on each side ranged from 10 to 15. The quantity and position of the dots were randomly assigned on every trial for each side separately, so that the display was never visually identical on both sides. The participants had to indicate which side of the screen contained more dots. This was a hard task because there was a maximum 5 point difference between the sides, and the number of dots on each side was continually adjusted based on the participants rate of success: If they had a high rate of success, the task was made harder by reducing the difference between the sides. After each trial, they were verbally told whether they had won or lost and either a smiling or angry face appeared on the screen. The emotional expression of the face was random so that sometimes, a win was accompanied by an angry face and a loss was accompanied by a smiling face, and vice versa.

Here’s what they found. There was greater activation of the striatum to positive feedback that was signaled by a smiling face, but this was reduced in participants with avoidant attachment. You can see that here with the downward sloping line: The participants with higher avoidance scores showed less activation of the striatum than participants with who were less avoidant. The researchers interpreted this as indicating that avoidant individuals were relatively impassive to social reward.

For anxiously attached individuals, a left amygdala response was evoked by angry faces associated with negative feedback, and correlated positively with anxious attachment. You can see that here with the upward sloping line: participants with high anxious scores showed more activation of the amygdala than participants who were less anxious. The researchers interpreted this finding as suggesting that anxiously attached individuals had an increased sensitivity to social punishment.

These results suggest that attachment styles have consequences even when socially interacting with strangers. In

To summarize, avoidant individuals have blunted responses in both subparts of the emotional mentalizing systems that signal social safety or threat and this reflects a deactivation of attachment needs. Avoidant attachment style also have a distinctive pattern of emotion regulation strategies: such individuals place greater reliance on suppression but have difficulties in using reappraisal to reduce the emotional charge of negative events.

In contrast, anxiously attached individuals exhibit enhanced responses to social emotional information signaling threat and aversion. They have a hyperactivation of attachment needs and subjective lack of behavioural control: They tend to exhibit high levels of emotional expressiveness, emotional dysregulation, worry, and impulsiveness in their relationships.

Vrtička, P., & Vuilleumier, P. (2012). Neuroscience of human social interactions and adult attachment style. Frontiers in Human Neuroscience, 6, 212.

Let’s switch gears now and take a look at oxytocin, a hormone that have been extensively studied in attachment formation.

Oxytocin reduces the amygdala’s response to fear-inducing stimuli and has a stress-reducing, anxiolytic effect. Heinrichs et al. (2003) administered either oxytocin or a placebo to a group of participants undergoing a stressful event. They had to prepare and deliver a speech to an unresponsive audience. In addition to the administration of oxytocin/placebo, they did or did not have social support from a best friend during the preparation phase. Both oxytocin and social support reduced the levels of cortisol associated with public speaking, and there were greater benefits when both occurred together than the effects of each alone. You can see that here with the open circles that refer to having both social support and oxytocin. And here it is in chart form. This may explain the known health-related advantages of having social support for a range of physical and mental health problems that tend to increase with stress.

Oxytocin enhances the formation of social and emotional bonds with others, even with strangers, by reducing stress and increasing trust. Within the context of group behaviour, oxytocin administration increases trust and cooperation with in-groups and a defensive, but not hostile, reaction to outgroups. We’ll look at groups more specifically in part 4.

Heinrichs, M., Baumgartner, T., Kirschbaum, C., & Ehlert, U. (2003). Social support and oxytocin interact to suppress cortisol and subjective responses to psychosocial stress. Biological Psychiatry, 54(12), 1389–1398.

Separation and social exclusion

Separation or rejection is often described as painful. To test whether it was really ‘pain’ Eisenberger, Lieberman, and Williams (2003) conducted an fMRI, or brain imaging, study of a cyberball game involving three players, including the person being scanned. Players could opt to throw the ball to one of the two other players. However, after a while, the game was fixed so that two players consistently threw to each other, excluding the person in the scanner. The anterior cingulate was activated by social exclusion and this activity was correlated with self-reported distress during the social exclusion. You can see that with the upward sloping line, where greater distress showed greater activation of the anterior cingulate. The anterior cingulate is known to respond to physical pain.

Eisenberger, N., Lieberman, M., & Williams, K. (2003). Does Rejection Hurt? An fMRI Study of Social Exclusion. Science, 302(5643), 290–292.

Other research shows that sensitivity to pain predicts sensitivity to social rejection on the cyberball game - that is, those who have a lower tolerance to thermal pain tend to report higher levels of unpleasant feelings following exclusion.

Grief has also been shown to activate the anterior cingulate. This is consistent with an overlap between social pain and physical pain found in the more artificial scenario of the cyberball game.

Opioids have a pain-killing, or analgesic, effect both for physical pain and for separation distress.

And a meta-analysis of findings in social psychology revealed that although social exclusion was associated with reported feelings of negative emotions (e.g. distress, sadness) there was a general trend for it to be accompanied by feelings of emotional blunting - for both positive and negative emotions. This is consistent with the notion that there may be an analgesic response to social exclusion, which results in a general emotional numbness.

These studies show that social exclusion may tend to rely on some of the same neural circuits as physical pain and may result in temporarily blunted emotions as a counter, or analgesic, response.

Blackhart, G. C., Nelson, B. C., Knowles, M. L., & Baumeister, R. F. (2009). Rejection elicits emotional reactions but neither causes immediate distress nor lowers self-esteem: A meta-analytic review of 192 studies on social exclusion. Personality and Social Psychology Review, 13, 269-309.

Negative social interactions

Perceived social isolation

Hypervigilance for social threats

Leads to biasesBehavioural confirmation

Social environment

Attraction and connection

Repulsion and isolation

Loneliness is a related concept to social exclusion but differs from it in a number of ways. It is more akin to a trait (a relatively stable disposition) than a state (induced by a particular situation). It can also differ from social exclusion in that it need not include an element of rejection by others. It could be an outcome of shyness or social anxiety, for instance. It could also reflect perceived isolation (such as a lack of intimacy) rather than to actual isolation. However, loneliness and social exclusion do tend to go together because lonely people often withdraw from social situations, the net result being that others do not include them in social activities or actively reject them. Lonely people may also look for rejection cues in other people and interpret ambiguous cues as threatening. This, of course, reinforces the feeling of loneliness.

Perceived loneliness has important health consequences. For example, it predicts elevated blood pressure, even after statistically removing the effects of objective social support and isolation. Cognitive decline in the elderly and risk of Alzheimer’s disease are predicted by perceived loneliness even after the amount of actual social contact and size of their social network is factored out. Note that these studies were correlational, meaning that the researchers found that these factors were related, not that loneliness caused the increased blood pressure or cognitive decline. Also, the participants in these studies were middle-aged or older. In one of the studies, the average age of the participants at the start of the study was about 80 years. Cognitive decline starts decades before it’s diagnosed so the participants who did go on to develop Alzheimer’s may have been in the early stages of Alzheimer’s disease at the start of the study.

Perceived loneliness can affect cognitive processing. When young, college aged, participants were told that their personality type meant that they would have a lonely future (this was not true), they experienced detrimental effects on intelligence measures, even though their current objective levels of social support were not altered by the manipulation. Specifically, cognitive performance in complex cognitive tasks such as effortful logic and reasoning declined whereas simple information processing remained intact despite the social exclusion. The effects were specific to social exclusion: participants who received predictions of future nonsocial misfortunes (accidents and injuries) performed well on the cognitive tests. The cognitive impairments were specific to the social isolation group and involved reductions in both speed (effort) and accuracy.

Baumeister, R. F., Twenge, J. M. & Nuss, C. K. (2002). Effects of social exclusion on cognitive processes: Anticipated aloneness reduces intelligent thought. Journal of Personality and Social Psychology, 83, 817-827.

Cole et al. (2007) found that loneliness affects gene expression relating to the body’s defense system from infection. More specifically, in lonely individuals the transcription of glucocorticoid response genes is impaired (here, isolated refers to high lonely individuals and integrated refers to low lonely individuals) and the activity of pro-inflammatory transcription control pathways is increased. What this means is that lonely individuals who experience chronically high levels of subjective social isolation have an elevated risk of inflammatory disease.

However, you should note that this study was at one point in time, and it doesn’t answer the question which came first. Did the social isolation cause the differences in gene transcription, or did the differences in gene transcription cause social isolation?

Cole, S. W., Hawkley, L. C., Arevalo, J. M., Sung, C. Y., Rose, R. M., & Cacioppo, J. T. (2007). Social regulation of gene expression in human leukocytes. Genome Biology, 8, 13.

Let’s now look at grief.

Grief is an intense feeling of loss that occurs as a result of permanent separation, normally death, from a loved one. O’Connor et al. (2008) conducted a brain imaging study on women who had lost a mother or sister due to breast cancer in the last five years. The grieving women consisted of two groups. One group was diagnosed with complicated grief in which there was no sign of abatement of their sense of loss, yearning for the loved one, or preoccupation with thoughts of them. The other group was classified as having non-complicated grief. Both groups were shown photographs of either the loved one or a stranger with grief-related words, such as ‘dying’, or neutral words superimposed on them. Both groups showed activity in a number of regions related to pain (dorsal anterior cingulate, insula, periaqueductal gray) when presented with images of the deceased compared to the stranger. This is consistent with an overlap between social pain and physical pain found in other studies. However, those with complicated grief activated the nucleus accumbens, seen here on the left, more than those with non-complicated grief. The nucleus accumbens is part of the reward network, so for those with enduring or complicated grief, looking at images of their deceased loved ones was rewarding. And this activity was correlated with subjective reports of yearning (you can see that here on the right with the upward sloping line). This was interpreted as evidence of an ongoing attachment to the deceased and an inability to fully adapt to their loss.

O’Connor, M. F., Wellisch, D. K., Stanton, A. L., Eisenberger, N. I., Irwin, M.R., & Lieberman, M. D. (2008). Craving love? Enduring grief activates brain’s reward center. Neuroimage, 42, 969-9732.

Queen Victoria experienced what clinical psychologists now call complicated grief following the early death of her husband, Prince Albert. For several years after his death she insisted that his personal effects should be laid out, such as hot water for shaving, and she appeared to be in a deep depression. She wore black for her remaining 40 years of reign. This photo was taken two years after her husband’s death in 1863 showing her in full mourning regalia.

Image by George Washington Wilson (1823-1893) - http://www.nationalgalleries.org/collection/online_search/4:324/result/0/63179, Public Domain, https://commons.wikimedia.org/w/index.php?curid=7109450

• Attachment

• Attachment style

• Oxytocin

• Social exclusion

Summary

An attachment is a long-enduring, emotionally meaningful tie to a particular individual, such as between mother and infant or between romantic partners. Love is the emotion that’s elicited in the presence of or by the thought of another person with whom there is an attachment. Love is associated with activity in the reward circuitry of the brain, the nucleus accumbens.

Attachment styles are partially conserved from infancy into adulthood. A person’s attachment history acts as a schema, or a pattern of thought, for predicting and interpreting social interactions whether intimate or not.

Oxytocin is important for forming attachment, enhancing social affiliation, and reducing stress.

Social exclusion, separation, and grief may tend to rely on some of the same brain circuits as physical pain. Lonely people have a tendency to process social stimuli differently, even when there is no element of social rejection.