Identity, Social Connection, and Loneliness

By: Curtis Peterson ©

This blog describes the historical development of the study of loneliness

Early conceptions of loneliness associated the experience of loneliness with more dispositional and personality qualities rather than as a part of normal social motivational processes. Additionally, according to early conceptualizations of loneliness, the experience of loneliness often leads to dysfunctional behaviors. Early focus on consequences of loneliness included study of the lonely housewife and cheating behaviors (Sells, 1948) or the lonely soldier drinking excessively and engaging in sexually promiscuous behaviors (Frosdick, 1918). Indeed, as will be indicated later in this chapter, individuals who experience severe levels of loneliness can lead to dysfunctional ways in alleviating that loneliness. The issue of these early conceptualizations of loneliness and consequences is that they were antidotal and were not measured by any objective means. As far as studies that directly address loneliness, according to the PsycINFO database, the earliest research was by Watson (1930) who looked at what makes educated individuals happy. In this exploration Watson found that loneliness was negatively associated with happiness, suggesting that loneliness was a dysfunctional process. Later in 1948, research by Thompson found that individuals who scored high on different psychosis scales on the ‘Minnesota Multphasic Personality Inventory’ (MMPI) also scored high on a subscale of loneliness. This lead Thompson to make similar conclusions as Watson in 1930 that loneliness was (a) a dysfunctional process and (b) that loneliness must somehow be more related to one’s disposition rather than the situational or social experiences of a person. The other aspect of Thompson’s research that will influence contemporary research is the notion that loneliness is closely related to depression, and is a key symptom. Indeed, current research has found a strong association between the experience of loneliness in one-time period (ex. middle childhood) and the development of depression in later time period (ex. adolescence) (Zimmer-Gembeck, Trevaskis, Nesdale, & Downey, 2014). Additionally, depressive symptoms tend to include analogous experiences of loneliness (Zimmer-Gembeck, Trevaskis, Nesdale, & Downey, 2014).

In contrast to Thompson (1948) where loneliness is seen as a feature of psychosis, Brooks (1933) asked rural psychiatrist’s whether social isolation (as defined as monotony and lonesomeness) was the cause of psychosis. While the results were mixed, Brooks concluded that social isolation is symptomatic of psychosis but does not cause psychosis. Rather, he concluded that poor socializing skills and coping skills that evolved from psychotic personality more likely lead to isolation and the experience of lonesomeness. This conclusion is used to explain the association between early experiences of loneliness and later experiences of depression (Jones, et al. 2011; Anderson, Miller, Riger, Dill, & Sedikides,1994; Cacioppo, Hawkley, & Thisted, 2010; Zimmer-Gembeck, Trevaskis, Nesdale, & Downey, 2014). For example, Jones, et al. (2011) found an association between loneliness in middle school children and adolescence’s experience of depression. These authors concluded that the experience of loneliness in middle childhood thwarted these children’s ability to develop socializing skills necessary to be a part of forming social relationships in adolescence leading to depression. The problem with Brooks (1933) and Jones, et al. (2011) conclusion is that they rely on descriptive and correlational methods, so it is purely theoretical on how early experiences of loneliness indeed predicts later development of depression.

The next stage in the development of the understanding of loneliness came from Sullivan in 1953 who was one of the first to define loneliness as a developmental – personality process. That is to say that loneliness is driven by the person’s disposition and attachment that occurs in early childhood. This idea of attachment and loneliness is still supported in today’s literature (Baumeister, & Leary, 1995), however, the notion of the lonely personality is not strongly supported (Hawkley, & Cacioppo, 2010). Rather loneliness is seen more as a need drive, similar to the need for food, where some individuals need a lot and some people need just a little to sustain their functioning (Cacioppo & Patrick, 2008). Probably the greatest legacy of Sullivan’s (1953) work is that it also provided a consistent definition of loneliness as “the need for intimate exchange with fellow being … with respect to satisfaction and security” (p. 261), the loss for which according to Sullivan causes “the driving of this system may integrate interpersonal situations despite severe anxiety.” (p. 262). In other words, the experience of loneliness creates a negative anxiety state which motivates an individual to relieve that negative state through interpersonal contact, despite the anxiety that accompanies the loneliness state. While this definition will be expounded on, specifically by Weiss (1973/1985), this definition remains the basic way psychologists have defined loneliness since.

The next evolution of the study of loneliness came in the 1970s with Weiss (1973/1985) book “Loneliness: The experience of Emotional and Social Isolation”. In this work Weiss argues that there are two types of loneliness. The first Weiss defined as emotional loneliness which is a negative affective state in which a person lacks close emotional ties with someone else. The second is social loneliness, or in Weiss’s terms social isolation, as a state of lacking sufficient social connections. Weiss used the term social isolation to emphasize the point that individuals who experience social isolation evaluate not having a sufficient number of social connections and social support in their life. Social isolation differs from emotional loneliness, in that emotional loneliness is the feeling of loss of close emotional relationship(s), whereas social isolation is a lack of sufficient social relationships, where close emotional or utility social support type relationships. Research since Weiss has supported the notion that there are two different types of loneliness, mainly from psychometric evidence (Vassar, & Crosby, 2008). The term social isolation continues to be used in contemporary research as defining social loneliness whereas the term loneliness often refers to emotional loneliness (Cacioppo, Cacioppo, & Capitanio, 2014). In addition to differentiating the two types of loneliness, Weiss (1973/1985) also introduced the idea that loneliness was not a dysfunctional process, but rather an ordinary motivational socializing processes. That is to say, according to Weiss, we are all driven to maintain a certain level of social and emotional connection and when our subjective experience goes below that individual threshold, we experience the negative state of loneliness. This, according to Weiss, drives us to either seek out emotional or social relationships. However, as will be presented in the contemporary research sections, individuals do not always go about seeking social connection in functional ways when experiencing loneliness (Hawkley, & Cacioppo, 2010).

In the 1980s and 1990s researchers became more and more interested in the reasons why individuals experience loneliness, beyond the subjective threshold, most likely driven by standardized measures such as the UCLA Loneliness Scale (Russel, Peplau, &Ferguson, 1978, Russel, 1996), which remains the most cited loneliness measurement scale used today (Ang, Mansor, & Tan, 2014). Since originally introduced, the UCLA Loneliness Scale has seen three revisions and development of two short forms of the measurement tool (Durak, & Senol-Durak, 2010). Additionally, the measure is the most translated scale on loneliness and has been translated and validated for populations from Zimbabwe (Wilson, Cutts, Lees, Mapungwana, & Maunganidze, 1992) to one of the most recent translations for the Turkish culture by Durak and Senol-Durak in 2010. The UCLA Loneliness Scale can be found in Appendix A.

With the advent of psychometric measures, some notable research that would influence the field up to current times were developed. From a cognitive perspective Conoley and Gerber (1985) investigated how loneliness affects reframing process of viewing the self and others. One of the more significant works that came out of the early 1990s was WcWhirter (1990), who provided one of the first reviews of the literature on loneliness, and the implications of counseling and research. Significant to WcWirter’s review was the presentation of data that indicated loneliness is a unique experience separate from any other dysfunctional state whether social (loss of loved one) or internal (depression or distressing experience) experiences. Indeed, WcWirter’s assertions have continued to be supported by current research (Hawkley, & Cacioppo, 2010). Other research worth noting of the early 1990s is the work of Lunt (1991), who continued the attributional work done by Conoley and Garber in 1985 and provided as causal model of loneliness based on attributional models. Using network and cluster analysis techniques, Lunt (1991) developed a thirteen variable model that clustered into five different groups. These clusters include: cluster one (physical unattractive and unpleasant personality), cluster two (others’ own groups-relationships, others’ lack of trying, and others’ fears), cluster three (impersonal situations and lack of opportunity), cluster four (lack of knowledge, lack of trying, shyness, and fear of rejection), and cluster five (pessimism and unlucky). While it is important to understand that these clusters developed not under causal experimental design, but by using psychometrics measures and self-report, they tend to support other self-report studies, but caution should be taken as this is not an experimental causal model but a psychometric – theoretical causal model. The significant outcome of Lunt’s (1991) study that is still an underlying assumption of the experience of loneliness today is the subjective cognitive evaluation. That is to say, loneliness suffered by an individual comes from a subjectively calculated estimate of experiencing of either loss of social connections or loss of close emotional connections with others. It is this evaluation that drives an individual attribution of to state of being, in Lunt’s work usually evaluated within one of the five clusters.

The 1990s study of loneliness saw a lot of attention and can be summarized in two late 1990 articles by Rockach and Brock (1997) on loneliness and life changes and Cacioppo and Gardner (1999) article entitled “Emotion”. By the late 1990s with the advent of more accessible use of computers and statistical software, more studies on loneliness using more advanced statistical methodology such as factorial analysis continued to garner support for a five factor model of loneliness that Lunt (1991) proposed. By the late 1990s, a five factor model which included (1) emotional distress, (2) social inadequacy and alienation, (3) growth and discovery, (4) interpersonal isolation, and (5) self-alienation seemed to be well support with-in the literature (Rockach & Brock, 1997). Work that supported this evidence and expanded on different dimensions of loneliness was that of Rockach and Brock in 1997.

Rockach and Brock (1997) investigated the five-factor model using a general population sample (versus traditional convenient sample of college students) that included 633 participants ranging in age from 13 to 87. In addition to using a general population sample, Rockach and Brock tested to see if there was variation among these five factors on five other variables that included (1) gender, (2) relationship status, (3) chronic or episodic loneliness, (4) current or past experience of loneliness, and (5) age at which loneliness is or was experienced. According to their findings, men’s experience of loneliness had greater loading on interpersonal isolation and perceived social alienation. Whereas women did not differ from the general population. The gender difference tends to be consistent with current findings with men experiencing loneliness more frequently, especially as they age (Victor & Bowling, 2012). For relationships, they found that individuals who were married experience loneliness the most intensely, and loaded heavily on growth and discovery, interpersonal isolation, and self-alienation. While consistent with current research (Segrin, Powell, Givertz, & Brackin, 2003) and research before Rockach and Brock (1997), divorced individuals experienced loneliness most frequently.

The third variable tested by Rockach and Brock (1997) was episodic versus chronic loneliness. According to these researchers by this time there was growing evidence that the experience of loneliness can lie on a continuum between individuals who tend to experience loneliness on a chronic basis to those who experience loneliness on an episodic level. At this time, chronic loneliness was seen as part of one’s personality, whereas episodic loneliness was event driven and was experienced by most of the population regardless of personality. The personality view of loneliness has since changed, as loneliness has been seen more as a natural driving force. Current researchers are finding evidence of more of a drive model analogous to food, where some people feel the need to consume more food than do other individuals, there are some individuals who have the need to find continued reductions in loneliness whereas others need minimal social and emotional contact to be satisfied (Lieberman, 2013).

Of interest in Rockach and Brock’s (1997) research was their focus on whether someone was currently experiencing loneliness, or if they were recalling past experiences of loneliness. Consistent with previous population rate research both in the 1990s (Lunt, 1991) and current (Cacioppo & Patrick, 2008), which suggests at any given time, about 25% of the population is experiencing loneliness, Rockach and Brock’s sample consisted of 30% slightly but not significantly above the population rate. According to their findings, the weight in which individuals who are currently experiencing loneliness significantly differ from individuals who were recalling episodes of loneliness in their past. Current lonely individuals weighted higher on social alienation, growth and discovery, and self-depreciation. With self-depreciation, the most salient aspects were social inadequacy and self-alienation, whereas individuals recalling episodes of loneliness weighted more heavily on emotional distress.

The most significant parts of the 1990s that contributed to the knowledge base on loneliness was the systematic investigations of the features of loneliness, represented in this review by the work of Lunt (1991) and Rockach and Brock (1997). The last article that will be reviewed before moving on to more contemporary issues are that of Cacioppo and Gardner (1999) on emotions. It should be noted that John Cacioppo has become one of the leading research experts in the field of loneliness and currently the director of the Center for Cognitive and Social Neuroscience at the University of Chicago where he has a robust research lab dedicated to the investigation of loneliness. One of the earlier works of Cacioppo, which included his colleague Wendi Gardner, was in 1999 in an article on emotions that emphasized three topics that would influence the study of emotions and loneliness (1) methodological issues, (2) relationship between emotions and cognition, and (3) the affective system which underlie emotions.

Methodological issues that were raised by Cacioppo and Gardner (1999) was for on advocating for the increased integration between the use of standardized measurements and neurological investigation tools such as fMRI imaging techniques. As will be seen in the last couple sections of this historical review, the last ten years has seen a robust increase and interest in the neurological processes associated with loneliness. By this time these authors’ argued that there had been developed several ways to measure emotions (i.e. self-report, indirect measures, and non-verbal measures) and that the field should continue the use and development of these tools. However, despite their advocacy for more lab-based studies, they also stated the field needs to address the ecological validity of the studies of emotions. In the context of loneliness, this call for more ecological validity will be seen in many of the contemporary studies that will be reviewed that investigates how loneliness is experienced within one’s community (Smith, 2012). Additionally, Cacioppo and Gardner (1999) argued that increased attention needed to be placed on socioemotional developmental aspects of loneliness. Again as will be indicated in the next section on contemporary issues, emphasis has been placed on socio-developmental processes of loneliness through childhood (Jones, Schinka, Van Dulman, Bossarte, & Swahn, 2011) to adulthood (Fokkema, Gierveld, & Dykstra, 2012).

The second area that Cacioppo and Gardner (1999) addressed was the relationship between cognition and emotions. They noted that that relationship between an individual’s emotion and an individual’s cognition (rationality) have been seen since the Greek philosopher times as an adversarial one with emotions being seen as blinding a person’s rational senses and abilities. However, as noted by Cacioppo and Gardner “[a]lthough the obstacles of a civilized world still occasionally call forth blind rage, emotions are increasingly recognized for the constructive role they play in higher forms of human experience” (p. 194). In other words, Cacioppo and Gardner argue that emotions have a key role in signaling that something must change and providing motivation to enact change, in a rational manner. Indeed, loneliness today is seen not as a self-defeating emotional process, but rather a negative emotional state that motivates the individual to seek out social connection in order to alleviate. While this may not be the case for some individuals who experience loneliness who try to alleviate it through dysfunctional means such as alcohol consumption, promiscuous sexual activity, or further social withdrawal leading to more loneliness and potential development of depressive symptoms, the majority of individuals’ loneliness leads to seeking out social and emotional connection with others (Hawkley, & Cacioppo, 2010).

In Cacioppo and Gardner’s (1999) paper they state “evolutionary forces do not value knowledge or truth per se but species survival” (p.198) and they go on to state when explaining the differentiation between hostile and threatening stimuli that “the human brain and body have been shaped by natural selection to perform this affective categorization and respond accordingly” (p.198). In the previous paragraph it was noted that emotional signals can aid in cognitive appraisal and increase the ability to make immediate and rational decisions. However, as can be seen by the statements just made, it is important not to overlook the evolutionary processes that underlie the affective system. This is the third point that Cacioppo and Gardner made in trying to understand and investigate emotions. There are two issues to the affective system that must be discussed in understanding the affective response of the individual, first is the learning process, in which affective experienced are shaped the classical conditioning and operant conditioning processes. The second is the understanding that emotions seem to have a two channel system one aimed at identifying and responding to threats and the other identifying and responding to safety and appetitive needs. When it comes to loneliness the threat channel may explain why individuals become more hypervigilant and weary of other’s intentions when feeling lonely (Lodder, Scholte, Clemens, Engels, Goosens, & Verhagen, 2015), whereas the safety and appetitive channel attempts to find a way to fulfill an individual social and social-emotional needs (Chang, et al., 2014). From a learning aspect, this maybe how individuals form poor habits when it comes to alleviating loneliness. For example, a person may realize that when they drink and become intoxicated they feel less threat and more social which alleviates their lonely state. Therefore, they learn that their loneliness can temporarily lowered through drinking. Cacioppo and Gardner’s (1999) article provided a framework for understanding the methodological, cognitive, affective variables that should be taken into account when studying emotions to the present day. Indeed, as will be review in the contemporary research section, researchers try to connect individuals experience of loneliness between neurological mechanism (methodology), cognitive system, and affective systems in order to understand how these systems respond to the environment and stimuli to try and understand the emotional state of loneliness.

The last evolution in the history of the study of loneliness fully developed within that last sixteen years with work from Cacioppo and Patrick (2008) in their book “Loneliness: Human Nature and the Need for Social Connection”, Hawkley, and Cacioppo (2010) in their article “Loneliness Matters: A Theoretical and Empirical Review of Consequences and Mechanisms” and Cacioppo, Cacioppo, and Capitanio (2014) in their article “Towards a Neurology of Loneliness” all of which describe the neurological basis of loneliness and the associated outcomes of prolonged loneliness. Additionally, they provide a framework for understanding loneliness as a part of an evolutionary advantage through social living, providing loneliness as a key negative emotional drive to engage a person in action when they experience insufficient amounts of social connection. These issues will be further explored in the contemporary section on loneliness in this chapter.

Before moving to more contemporary research two issues must be explored, the first is the difference between loneliness as defined by Weiss (1973/1985) and loneliness that is experienced when one has depression. The second is the difference between the experience of loneliness and the experience of social rejection. Loneliness and depression seem to be intimately intertwined as loneliness is a feature of depression (Zimmer-Gembeck, Trevaskis, Nesdale, & Downey, 2014). However, being lonely does not necessarily means one is depressed (Cicognani, Klimstra, & Goossens, 2014).

Loneliness as a property of depression is not new, however as can be seen in this review, depression and loneliness have a long standing relationship. What seems to differ between normative loneliness (i.e. loneliness that experienced whenever there is a discrepancy in social connection) and depressive loneliness, seems to be the motivational state. That is, individuals who experience loneliness while depressed tend to withdraw from opportunities to regain social connection whereas individuals who are experiencing normative loneliness have a motivation to alleviate the negative emotions it produces through seeking out social opportunities (de Minzi, 2006). Why loneliness has a different effect when a person is depressed versus not depressed is still under investigation. Research by Segrin, Powell, Givertz, and Brackin (2003) did research on didactic couples who one or both were experiencing depression suggests that loneliness becomes part of the negative affective-cognitive rumination cycle, and therefore instead of seeking out social connection the negative emotions associated with loneliness reinforces the person’s belief, attitudes, and behaviors, including their lack of abilities to create meaningful social connections.

From Segrin, Powell, Givertz, and Brackin (2003) research one can glean processes that are present when experiencing loneliness in that it includes a negative affective state (e.g. feeling anxious), a cognitive state (e.g. appraisal of social situation), and behavioral/motivational state (e.g. seeking out social connection). From Segrin, Powell, Givertz, and Brackin (2003) research it seems that depression interrupts the motivational and behavioral process (ex. withdraw from social situations) by creating a different evaluation of the person affective state (ex. anxious because lack of relationships) and cognitive appraisal (ex. I must be anxious because I am not good at relationships).

A second distinction that needs to be made before moving on is the distinction between social rejection and loneliness. Mainly, how do these two experiences differ from one another or do they? The best evidence for the differences between social rejection and loneliness come from neurological studies using fMRI to measure different activation during tasks where individuals are lonely versus not lonely and other studies that measure brain activity when someone experiences some type of social rejection. Research of this type has indicated that different brain regions become active when experiencing rejection versus being in a lonely state. For example, research by Eisenberger, Lieberman, and Williams (2003) and Eisenberger (2012) had individuals experience simulated social rejection while in an fMRI machine. This activity known as “cyber ball” requires a person to pass a ball on a video screen to one of two other players. In the social rejection scenario, the two other players stop passing the ball to the participant. When measuring activation in the brain, the researchers found that more the person experience social distress (i.e. rejection) the more of the area of the brain known as the dorsal anterior cingulate cortex (dACC) activated. Interestingly the dACC is also activated when a person experiences physical pain. However, similar research done on lonely individuals develops a different pattern. Instead, area associated with motivation (ventral striatum, caudate nucleus, and temporal gyrus) and emotions (amygdala, thalamus, and hypothalamus) become active under different lonely stimuli conditions (Cacioppo, Norris, Decety, Monteleone, & Nubaum, 2008; Cacioppo, Cacioppo, & Capitanio, 2014).

These studies potentially highlight the difference between rejection – as pain – and loneliness – as motivational emotion, there should be some caution in interpreting such studies. First they compare different regions of the brain under different types of procedures and activities. Second and most importantly, in the rejection condition individuals experience the state of rejection while having their brain scanned. Whereas as in the loneliness condition this is a pre-disposed state in which comparisons between individuals who scored low on a loneliness scale are compared to individuals who score high on a loneliness scale. Therefore, while these studies provide starting evidence that there is a distinction between rejection and loneliness, there is much more work that needs to be done in this area. Additionally, independent replications of Eisenberger, Lieberman, and Williams (2003) and Eisenberger (2012) work have failed (Cacioppo, et al. 2013). This indicates that while there may be a differentiation between social rejection and loneliness based on neural pathways, much more work needs to be done to identify exactly what the differences are and how they manifest themselves in the neural system.