The behavioral, anatomical and pharmacological parallels between social attachment, love and addiction

Abstract

Rationale: Love has long been referred to as an addiction in literature and poetry. Scientists have often made comparisons between social attachment processes and drug addiction, and it has been suggested that the two may share a common neurobiological mechanism. Brain systems that evolved to govern attachments between parents and children and between monogamous partners may be the targets of drugs of abuse and serve as the basis for addiction processes.

Objectives: Here, we review research on drug addiction in parallel with research on social attachments, including parent-offspring attachments and social bonds between mating partners. This review focuses on the brain regions and neurochemicals with the greatest overlap between addiction and attachment and, in particular, the mesolimbic dopamine (DA) pathway.

Results: Significant overlap exists between these two behavioral processes. In addition to conceptual overlap in symptomatology, there is a strong commonality between the two domains regarding the roles and sites of action of DA, opioids, and corticotropin-releasing factor. The neuropeptides oxytocin and vasopressin are hypothesized to integrate social information into attachment processes that is not present in drug addiction.

Conclusions: Social attachment may be understood as a behavioral addiction, whereby the subject becomes addicted to another individual and the cues that predict social reward. Understandings from both fields may enlighten future research on addiction and attachment processes.

Fig. 1

The role of DA in pair bonding. (A) Prairie voles that mate with a partner during a 24-hour cohabitation will spend more time with the partner rather than a stranger during a subsequent partner preference test, which is the operational definition of a pair bond. Prairie voles injected with either saline or a DA D1R antagonist prior to this preference test form a pair bond normally, while injection of a D2R antagonist prior to cohabitation prevents pair bonding. These antagonist effects were shown in subsequent experiments to occur in the NAC shell. (B) Prairie voles that spend 6 hours with a partner without mating do not form a pair bond with the partner. Microinjection of a D2R agonist induces the formation of a pair bond during this 6-hour cohabitation, an effect that is blocked by a D2R antagonist. An OT antagonist also blocks agonist-induced pair bonding, suggesting that concurrent activation of D2R and OTR is necessary for pair bonding. Figure adapted from Young and Wang 2004.

Fig. 2

MOR and pair bonding. (A) Receptor autoradiography showing ligand binding to MOR in prairie vole brain. MOR density in the NAC shell is moderate, but much lower than MOR density in the CP. (B) After 24 hours of cohabitation with a partner, prairie voles receiving saline to the NAC shell or CP formed pair bonds as normal. MOR antagonist injected into the NAS shell did not affect pair bonding, while MOR antagonist in the CP prevented the formation of a pair bond. These data show that MOR in the CP is necessary for pair bond formation. Figure adapted from Burkett et al 2011.

Fig. 3

OT and AVP in pair bonding. (A) Monogamous prairie voles have high densities of OTR in the PFC, CP, and NAC. (B) By comparison, non-monogamous montane voles have relatively low densities of OTR in these regions. (C) Infusion of an OTR antagonist into the PFC and NAC of prairie voles, but not the CP, during a 24-hour cohabitation prevents the formation of pair bonds. (D) Male prairie voles have a higher density of AVP V1aR in the VP than do (E) montane voles. Infusion of a V1aR antagonist into the VP, but not into the mediodorsal thalamus (MDThal) or medial amygdala (MeA), of male prairie voles during a 24-hour cohabitation prevents the formation of pair bonds. This suggests that species differences in OTR and V1aR density in these regions may be a direct causal factor in species differences in social attachment. Figure adapted from Young and Wang 2004.

Fig. 4

Overlapping circuits for attachment and addiction. The VTA sends dopaminergic projections to the NAC, prefrontal cortex (PFC), and CP; these projections are all implicated in addiction, while only projections to NAC are implicated in attachment. The extended amygdala (E-Amg) is the presumptive source of AVP to the VP and LS in attachment, and CRF and glutamate to the NAC in addiction and attachment. The PVN is the source of OT release in the E-Amg and NAC. Glutamatergic projections link the PFC with the NAC, and GABAergic projections link the NAC and VP.