Love

Abstract

Love, which MeSH classifies under emotions, is the affective bond and set of feelings that draw one person toward and hold them to another, studied in psychology chiefly as romantic and parental love. This article covers whether love is an emotion, a motivation or an attachment process; the triangular, love-styles, passionate-love and attachment accounts; the scales that measure love and what meta-analysis finds they share; the reward-system and attachment neuroscience of romantic and maternal love; and current work on love regulation and the cross-cultural structure of love. The central dispute is whether love is one thing at all, or a single word for several systems with different causes and time courses. Three interactive demos show the triangular theory's components, what a moderate correlation between love and liking looks like, and the reported effects of regulating love feelings.

Keywords: love, romantic love, attachment, passionate love, triangular theory

Love entered experimental psychology late. When Rubin introduced a social-psychological love scale in 1970, he found it surprising that social psychologists had devoted virtually no attention to love, and noted that for Heider loving was merely intense liking (Rubin, 1970). Four decades on, a review observed that psychologists' efforts to understand love began in the mid-twentieth century, that they continue apace, and that an understanding of love has yet to be achieved (Berscheid, 2010). The same review names the failure to explicate the love construct as the major obstacle to the study of love phenomena (Berscheid, 2010). Much of what follows is an account of the competing explications.

What Love Is

The construct has been defined in three ways. The first treats love as an attitude: Rubin began from the assumption that love is an interpersonal attitude and built a scale whose conception of romantic love included three components, affiliative and dependent need, a predisposition to help, and an orientation of exclusiveness and absorption (Rubin, 1970). The second treats it as a structure of components: in the triangular theory, love has intimacy, which encompasses feelings of closeness, connectedness and bondedness; passion, which encompasses the drives that lead to romance, physical attraction and sexual consummation; and decision/commitment, the short-term decision that one loves another and the long-term commitment to maintain that love (Sternberg, 1986). The third treats it as a bond: romantic love was proposed to be an attachment process, a biosocial process by which affectional bonds are formed between adult lovers just as they are formed earlier in life between infants and their parents (Hazan & Shaver, 1987).

A broader working definition comes from evolutionary biology. One integrative review organises what is known about romantic love by Tinbergen's four questions, mechanism, development, function and phylogeny, contends that romantic love is a suite of adaptations and by-products, and proposes a new ethologically informed working definition to accommodate a multi-faceted understanding (Bode & Kushnick, 2021). The same review reports that romantic love is known to be present in almost all human societies (Bode & Kushnick, 2021). These definitions answer different questions, about what love feels like, what it is made of, and what it is for, and they need not compete.

The word itself has also been studied as a concept. A series of six studies found that the natural language concept of love is better understood from a prototype than a classical perspective: it has an internal structure and fuzzy borders, so that maternal love, romantic love, affection, love of work, self-love, infatuation and other subtypes can be reliably ordered from better to poorer examples of love (Fehr & Russell, 1991). Each subtype's goodness as an example of love predicted various indices of its cognitive processing, and the authors draw implications for a scientific definition and typology of love (Fehr & Russell, 1991).

Is Love an Emotion?

Emotion theory long left love out. One analysis documents the neglect of love in many contemporary emotion theories despite its prominence in the lay psychology of emotion, and argues that love should be considered a basic emotion like anger, sadness, happiness and fear (Shaver et al., 1996). Its resolution of the dispute is a distinction: the momentary surge form of love is a basic emotion with properties similar to joy, sadness and fear, whereas relational love is a bond that develops between people and is associated with many other emotions, such as distress and anxiety (Shaver et al., 1996). On this account love is the broad everyday name for emotions related to three interrelated behavioural systems, attachment, caregiving and sex (Shaver et al., 1996).

One test of the emotion claim asks whether love has a signal. In interactions that elicited love and threatened the bond, four affiliation cues, head nods, Duchenne smiles, gesticulation and forward leans, correlated with self-reports and partner estimates of love, which the authors take as evidence for a nonverbal display of love (Gonzaga et al., 2001). The experience and display of love also correlated with commitment-enhancing processes, such as constructive conflict resolution and perceived trust, when the relationship was threatened (Gonzaga et al., 2001). A rival reading emphasises motivation: early-stage romantic love has been described as possibly a developed form of a mammalian drive to pursue preferred mates (Aron et al., 2005). The two readings meet in the claim that romantic love is a commitment device for motivating pair-bonding (Fletcher et al., 2015).

Love is also distinct from desire. Sexual desire and romantic love are often experienced in concert but are fundamentally distinct subjective experiences with distinct neurobiological substrates, because the processes underlying sexual desire evolved in the context of sexual mating while those underlying romantic love originally evolved in the context of infant-caregiver attachment (Diamond, 2004). The fuller statement of this view is a biobehavioral model specifying that the evolved processes underlying sexual desire and affectional bonding are functionally independent, that the processes underlying affectional bonding are not intrinsically oriented toward other-gender or same-gender partners, and that the links between love and desire are bidirectional, particularly among women (Diamond, 2003). One consequence is that an individual's sexual predisposition may not circumscribe their capacity to fall in love with partners of either gender (Diamond, 2004); the distinction is developed further on the libido page.

Theories of Love

The triangular theory is a prominent structural account. The amount of love one experiences depends on the absolute strength of the three components, and the kind of love depends on their strengths relative to each other; the components interact with each other and with the actions they produce, so as to form a number of different kinds of loving experiences (Sternberg, 1986). The theory is claimed to subsume certain other theories and to account for a number of empirical findings in the research literature (Sternberg, 1986).

The love-styles account classifies love by attitude rather than component. It extends Lee's theory of six basic love styles: Eros (passionate love), Ludus (game-playing love), Storge (friendship love), Pragma (logical, shopping-list love), Mania (possessive, dependent love) and Agape (all-giving, selfless love) (Hendrick & Hendrick, 1986). Lee had constructed the styles as ideal types derived from fictional and non-fictional literature, devised an interview method that let individuals in love reconstruct their experience in codeable form, and tested and revised the types into a structured typology of readily distinguishable styles of loving (Lee, 1977). In a 42-item questionnaire completed by 807 respondents, six love-style scales emerged clearly from factor analysis with low intercorrelations, and a confirmatory study of 567 respondents replicated the factor structure (Hendrick & Hendrick, 1986). Love attitudes were related to gender, previous love experiences, current love status and self-esteem (Hendrick & Hendrick, 1986).

Passionate love has its own measure. In developing the Passionate Love Scale, its authors defined passionate love, explained the need for a scale to measure it, and presented evidence of its reliability, validity and relationship to other factors in close relationships (Hatfield & Sprecher, 1986). They describe using the scale in family therapy to open conversations about passionate love, companionate love and intimacy (Hatfield & Sprecher, 1986).

The attachment account makes love developmental. Its authors translated key components of attachment theory, developed to explain the development of affectional bonds in infancy, into terms appropriate to adult romantic love, centring on the three major infant attachment styles, secure, avoidant and anxious/ambivalent, with continuity of relationship style due in part to mental models of self and social life (Hazan & Shaver, 1987). Two questionnaire studies indicated that the relative prevalence of the three styles is roughly the same in adulthood as in infancy and that the three kinds of adults differ predictably in how they experience romantic love (Hazan & Shaver, 1987). Measurement has since moved on: an item response theory analysis of four commonly used self-report attachment inventories, in data from 1,085 individuals, warned that scoring them by averaging or summing items can lead to erroneous inferences about issues such as the continuity of attachment security, and found that commonly used attachment scales can be improved in a number of important ways (Fraley et al., 2000).

A temporal account cuts across all four. A review of the field suggests that it is important to consider the presence or absence of four types of love, each apparently associated with different causal conditions and therefore likely to have a different temporal course as an adult relationship moves through time (Berscheid, 2010). The evolutionary account adds a three-system view: the sex drive evolved to motivate individuals to seek a range of mating partners, attraction to motivate them to prefer and pursue specific partners, and attachment to motivate them to remain together long enough to complete species-specific parenting duties (Fisher et al., 2006). That account now has a named rival. For 25 years the predominant evolutionary theory of romantic love has been Fisher's theory of independent emotion systems, which holds that sex drive, romantic attraction and attachment are associated with distinct neurobiological and endocrinological systems that evolved independently of each other; a competing theory proposes that co-opting mother-infant bonding sometime in the recent evolutionary history of humans may partially account for the evolution of romantic love (Bode, 2023).

Figure 1

Three Behavioural Systems in the Evolutionary Account of Romantic Love

Measuring Love

Measurement began with a distinction between loving and liking. Rubin's love-scale scores were only moderately correlated with scores on a parallel scale of liking, which reflected a more traditional conception of interpersonal attraction (Rubin, 1970). The validity test was behavioural: college dating couples who loved each other a great deal, by their love-scale scores, were predicted to spend more time gazing into one another's eyes than couples who loved each other less, and the prediction was confirmed (Rubin, 1970).

Many scales followed, and meta-analysis has asked what they share. Correlations between several commonly used measures of love, aggregated across 81 studies representing 103 samples and 19,387 individuals, yielded factors suggesting general love, romantic obsession and practical friendship (Graham, 2011). Love was positively and obsession negatively associated with relationship satisfaction and length, and the author concluded that the field needs measurement methods other than global self-report (Graham, 2011). The cross-cultural question has been asked of the triangular scale: in a final sample of 7332 participants from 25 countries on all inhabited continents, tests of configural, metric and scalar invariance confirmed the cultural universality of the construct as measured, and levels of the love components differed with relationship duration in the way the triangular theory suggests (Sorokowski et al., 2021).

Table 1

Selected Measures of Love, as Described in the Cited Abstracts

Note. Each row lists only what the cited paper's abstract reports (Rubin, 1970; Hendrick & Hendrick, 1986; Hatfield & Sprecher, 1986; Sorokowski et al., 2021); the components of the triangular scale are those of the theory (Sternberg, 1986). The higher-order factors underlying the most popular measures of love are identified meta-analytically by Graham (2011).

The Neuroscience of Love

The imaging literature began with one study. When 17 people who were deeply in love viewed pictures of their partners rather than of three friends, activity was restricted to foci in the medial insula and anterior cingulate cortex and, subcortically, in the caudate nucleus and putamen, with deactivations in the posterior cingulate gyrus and amygdala (Bartels & Zeki, 2000). The authors suggested that a unique network of areas is responsible for evoking this affective state (Bartels & Zeki, 2000). A follow-up compared maternal with romantic love: both activated regions specific to each as well as overlapping regions in the brain's reward system that coincide with areas rich in oxytocin and vasopressin receptors, and both deactivated a common set of regions associated with negative emotions, social judgement and mentalizing (Bartels & Zeki, 2004). The authors concluded that human attachment employs a push-pull mechanism that overcomes social distance by deactivating networks used for critical social assessment while bonding individuals through the reward circuitry (Bartels & Zeki, 2004).

The reward reading was then tested directly. In 10 women and 7 men who had been intensely in love for 1 to 17 months, activation specific to the beloved occurred in dopamine-rich areas associated with mammalian reward and motivation, namely the right ventral tegmental area and the right postero-dorsal body and medial caudate nucleus, and activation in the right anteromedial caudate correlated with questionnaire scores for the intensity of romantic passion (Aron et al., 2005). The authors suggested that romantic love uses subcortical reward and motivation systems to focus on a specific individual (Aron et al., 2005). In 15 men and women who had just been rejected in love, a preliminary analysis showed activity specific to the beloved in reward regions associated with gambling for uncertain large gains and losses (Fisher et al., 2006).

Long marriages were examined next. In 10 women and 7 men married an average of 21.4 years, effects specific to the intensely loved partner were found in dopamine-rich reward and basal ganglia regions, such as the ventral tegmental area and dorsal striatum, consistent with early-stage studies, and in several regions implicated in maternal attachment, such as the globus pallidus, substantia nigra and raphe nucleus (Acevedo et al., 2012). The authors conclude that for some individuals the reward value associated with a long-term partner may be sustained, similar to new love, but also involves brain systems implicated in attachment and pair-bonding (Acevedo et al., 2012); the basal ganglia are the common thread.

The neurochemistry comes largely from animals. Studies of pair bonding in monogamous rodents such as prairie voles offer the most accessible animal models for the mechanisms underlying selective social attachments, and oxytocin and vasopressin have been implicated both in social bonding and in central control of the hypothalamic pituitary adrenal axis (Carter, 1998). A recent overview describes oxytocin and dopamine interacting to link the neural representation of partner stimuli with the social reward of courtship and mating, and vasopressin facilitating mate-guarding behaviours, potentially related to the human experience of jealousy (Blumenthal & Young, 2023). A conceptual framework integrating animal research with neuroimaging describes four mammalian bonds, parent-infant, pair-bonds, peers and conspecifics, all built upon systems shaped by maternal provisions during sensitive periods, with bonding underpinned by crosstalk of oxytocin and dopamine in the striatum (Feldman, 2017). The bonding mechanism itself is treated on the pair bond page.

Worked Example

The sample sizes in this literature repay arithmetic. The love-measures meta-analysis pooled 19,387 individuals across 103 samples (Graham, 2011), an average of about 188 per sample; the triangular-scale study analysed 7332 participants from 25 countries (Sorokowski et al., 2021), about 293 per country. Against those, four of the imaging samples cited here each numbered 17 people or fewer: 17 deeply in love (Bartels & Zeki, 2000), 17 early-stage lovers (Aron et al., 2005), 17 long-married participants (Acevedo et al., 2012) and 15 recently rejected (Fisher et al., 2006). The questionnaire samples are therefore roughly 11 to 20 times larger per sample than the imaging samples, and several hundred to over a thousand times larger in total.

The durations are as instructive. The early-stage participants had been in love for 1 to 17 months (Aron et al., 2005); the married participants had been married an average of 21.4 years, or about 257 months (Acevedo et al., 2012), roughly 15 times the longest early-stage duration. That both groups showed ventral tegmental activity is what licenses the claim that, for some individuals, the reward value of a partner may be sustained across that span (Acevedo et al., 2012).

Discussion

Three lessons follow for cognitive psychology. First, Shaver et al. argue that a number of controversies over whether love is an emotion can be resolved once momentary and relational love are separated (Shaver et al., 1996). On their account the surge form of love fits the criteria for a basic emotion, while relational love is a bond accompanied by many emotions (Shaver et al., 1996); a nonverbal display correlated with self-reported love (Gonzaga et al., 2001); and the drive-like profile of early romantic love in reward regions (Aron et al., 2005) describes a motivational state. The review that names the unexplicated construct as the field's main obstacle (Berscheid, 2010) is, on this reading, a call to stop asking one question of several phenomena.

Second, the structural theories share some of their distinctions. The meta-analysis of popular measures recovered general love, romantic obsession and practical friendship (Graham, 2011); the love-styles account separates passionate, possessive and friendship love among its six styles (Hendrick & Hendrick, 1986); and the triangular theory separates intimacy and passion (Sternberg, 1986). Obsession was negatively associated with satisfaction and length while love was positively associated (Graham, 2011), so not every factor underlying love measures relates to a relationship in the same direction.

Third, several neuroscience findings are consistent with the attachment account. Overlapping reward regions rich in oxytocin and vasopressin receptors were activated by both maternal and romantic love (Bartels & Zeki, 2004); long-term love recruited regions implicated in maternal attachment (Acevedo et al., 2012); and romantic love's processes are argued to have evolved in the context of infant-caregiver attachment (Diamond, 2004). The proposal that romantic love is an attachment process (Hazan & Shaver, 1987) was made before any of this imaging existed, and the evolutionary proposal that romantic love co-opted mother-infant bonding is offered explicitly as a competitor to the theory of independently evolved systems (Bode, 2023).

Current Directions

The research front runs in three directions. The first is regulation: explicit love regulation, defined as the use of behavioural and cognitive strategies to change the intensity of current feelings of romantic love, has been proposed and tested, with reported strategies including cognitive reappraisal, distraction, avoidance and undertaking new activities (Langeslag & van Strien, 2016). Instructed up-regulation using reappraisal increased subjective feelings of attachment and down-regulation decreased infatuation and attachment, although the event-related potential results were complicated by up-regulation also reducing the late positive potential between 700 and 3000 ms in participants who were involved in a relationship (Langeslag & van Strien, 2016).

The second is universality. The triangular scale's invariance across 25 countries is offered as confirming the cultural universality of the construct, and its data and adapted versions in a few dozen languages are intended to enable more extensive research (Sorokowski et al., 2021). The third is evolution: a review of romantic love under Tinbergen's four questions concludes that romantic love probably evolved in concert with pair-bonds in recent human ancestors, advances four potential scenarios for its evolution, and highlights outstanding questions as avenues for future research (Bode & Kushnick, 2021). The commitment-device thesis meets its challenges in the existence of arranged marriage, polygyny, divorce and infidelity (Fletcher et al., 2015), and a conceptual framework of human attachment now integrates insights from animal research with neuroimaging studies (Feldman, 2017).

Common Misconceptions

“Love is just intense liking.”

Rubin built separate love and liking scales and found their scores only moderately correlated (Rubin, 1970). For Heider, an earlier theorist of interpersonal attraction, loving was merely intense liking (Rubin, 1970).

“Romantic love and sexual desire are the same thing.”

The two are often experienced in concert but are fundamentally distinct subjective experiences with distinct neurobiological substrates and different evolutionary origins (Diamond, 2004).

“The reward of romantic love always fades after the first few years.”

In people married an average of 21.4 years who reported intense love, viewing the partner activated dopamine-rich reward regions consistent with early-stage studies, suggesting that for some individuals the reward value of a long-term partner may be sustained (Acevedo et al., 2012).

“Love feelings cannot be controlled.”

People do perceive love feelings as somewhat uncontrollable, yet instructed reappraisal increased reported attachment when used to up-regulate love and decreased infatuation and attachment when used to down-regulate it (Langeslag & van Strien, 2016).

Glossary

Agape.

In the love-styles account, all-giving, selfless love.

Attachment style.

In Hazan and Shaver's original account, one of three patterns, secure, avoidant and anxious/ambivalent, first described in infancy and translated into terms for adult romantic love.

Caudate nucleus.

A subcortical structure active when people in love view their beloved; activity in its anteromedial part correlated with the intensity of romantic passion.

Decision/commitment.

In the triangular theory, the short-term decision that one loves another and the long-term commitment to maintain that love.

Eros.

In the love-styles account, passionate love.

Intimacy.

In the triangular theory, the feelings of closeness, connectedness and bondedness experienced in loving relationships.

Love regulation.

The use of behavioural and cognitive strategies to change the intensity of current feelings of romantic love.

Ludus.

In the love-styles account, game-playing love.

Mania.

In the love-styles account, possessive, dependent love.

Pair bond.

A selective social attachment between mates, studied most accessibly in monogamous rodents such as prairie voles.

Passion.

In the triangular theory, the drives that lead to romance, physical attraction and sexual consummation.

Pragma.

In the love-styles account, logical, shopping-list love.

Relational love.

A bond that develops between people, associated with surge love and many other emotions such as distress and anxiety.

Romantic obsession.

One of three higher-order factors underlying popular love measures, negatively associated with relationship satisfaction and length.

Storge.

In the love-styles account, friendship love.

Surge love.

The momentary form of love, proposed as a basic emotion with properties similar to joy, sadness and fear.

Ventral tegmental area.

A dopamine-rich region active when people intensely in love view the beloved partner, both early in a romance and in some long marriages.

Key Researchers

Arthur Aron

. Research Professor of Social and Health Psychology at Stony Brook University; social psychologist of close relationships who led the functional imaging study of reward, motivation and emotion systems in early-stage intense romantic love. Wikipedia

Ellen Berscheid

(1936-2025). Regents Professor of Psychology at the University of Minnesota; a pioneer of the scientific study of interpersonal attraction and close relationships and author of a review of psychology's research on love. Wikipedia - Faculty Page

Helen E. Fisher

(1945-2024). Senior Research Fellow at the Kinsey Institute, Indiana University; biological anthropologist who proposed that romantic love is a mammalian brain system for mate choice. Wikipedia

Elaine Hatfield

. Professor of Psychology at the University of Hawaii at Manoa; with Susan Sprecher, developed the Passionate Love Scale for measuring passionate love in intimate relationships. Wikipedia - Faculty Page

Cindy Hazan

. Member of the Department of Psychology at Cornell University; with Phillip Shaver, conceptualized romantic love as an attachment process. Wikipedia - Faculty Page

Sandra J. E. Langeslag

. Associate Professor of Psychological Sciences at the University of Missouri-St. Louis, where she directs the Neurocognition of Emotion and Motivation Lab; studies whether and how romantic love feelings can be regulated. Wikipedia - Faculty Page

Phillip R. Shaver

. Distinguished Professor Emeritus of Psychology at the University of California, Davis; extended attachment theory to adult romantic relationships with Cindy Hazan and asked whether love is a basic emotion. Wikipedia - Faculty Page

Robert J. Sternberg

. Professor of Psychology in the College of Human Ecology at Cornell University; proposed the triangular theory of love, in which love comprises intimacy, passion and decision/commitment. ORCID - Wikipedia - Google Scholar - Faculty Page

Semir Zeki

. Professor of Neuroesthetics at University College London; with Andreas Bartels, reported functional imaging studies of the neural basis of romantic and maternal love. ORCID - Wikipedia - Google Scholar - Faculty Page

Frequently Asked Questions

What is love in psychology?

In one prominent account, the triangular theory, love has three components, intimacy, passion and decision/commitment (Sternberg, 1986).

Is love an emotion?

One influential answer distinguishes the momentary surge form of love, proposed as a basic emotion like joy, sadness and fear, from relational love, a bond that develops between people and is associated with many other emotions (Shaver et al., 1996).

What is the difference between love and liking?

Rubin built separate scales for the two and found that love-scale scores were only moderately correlated with liking-scale scores, which reflected a more traditional conception of interpersonal attraction (Rubin, 1970).

What are the six love styles?

The love-styles account names Eros (passionate love), Ludus (game-playing love), Storge (friendship love), Pragma (logical love), Mania (possessive, dependent love) and Agape (all-giving, selfless love), and six scales for them emerged clearly from factor analysis (Hendrick & Hendrick, 1986).

How is romantic love related to attachment?

Romantic love has been proposed to be an attachment process by which affectional bonds form between adult lovers as they form between infants and parents, and two questionnaire studies indicated that the relative prevalence of secure, avoidant and anxious/ambivalent styles is roughly the same in adulthood as in infancy (Hazan & Shaver, 1987).

What happens in the brain when people are in love?

Viewing a beloved partner activated dopamine-rich areas associated with reward and motivation, the right ventral tegmental area and parts of the caudate nucleus, in people intensely in love for 1 to 17 months (Aron et al., 2005).

Can romantic love last in a long marriage?

In people married an average of 21.4 years, viewing an intensely loved partner activated reward regions seen in early-stage love and regions implicated in maternal attachment, suggesting that for some the reward value of a long-term partner may be sustained (Acevedo et al., 2012).

Can people control their feelings of love?

People perceive love feelings as somewhat uncontrollable, but in the first studies of love regulation instructed reappraisal increased reported attachment when used to up-regulate love and decreased infatuation and attachment when used to down-regulate it (Langeslag & van Strien, 2016).

References