Reconsidering anhedonia in depression: lessons from translational neuroscience

Abstract

Anhedonia is a core symptom of major depressive disorder (MDD), the neurobiological mechanisms of which remain poorly understood. Despite decades of speculation regarding the role of dopamine (DA) in anhedonic symptoms, empirical evidence has remained elusive, with frequent reports of contradictory findings. In the present review, we argue that this has resulted from an underspecified definition of anhedonia, which has failed to dissociate between consummatory and motivational aspects of reward behavior. Given substantial preclinical evidence that DA is involved primarily in motivational aspects of reward, we suggest that a refined definition of anhedonia that distinguishes between deficits in pleasure and motivation is essential for the purposes of identifying its neurobiological substrates. Moreover, bridging the gap between preclinical and clinical models of anhedonia may require moving away from the conceptualization of anhedonia as a steady-state, mood-like phenomena. Consequently, we introduce the term "decisional anhedonia" to address the influence of anhedonia on reward decision-making. These proposed modifications to the theoretical definition of anhedonia have implications for research, assessment and treatment of MDD.

Copyright © 2010 Elsevier Ltd. All rights reserved.

Figures

Figure 1

Schematic illustration of dopamine projection pathways and circuitry regulating DA release in the human brain. DA firing rates are maintained at tonic levels in part due to steady-state inhibitory firing from the ventral pallidum. Excitatory projections from prefrontal cortex project, amygdala and hippocampus synapse on striatal targets, including the nucleus accumbens. The nucleus accumbens sends GABAergic projections to the ventral pallidum, suppressing VP inhibition of VTA, thereby facilitating phasic burst-firing of VTA DA neurons. Note: Placement of structure labels is approximate. Amyg = amygdala; Caud = Caudate; DA = Dopamine; GABA = GABAergic projections; Glu = glutamatergic projections; Hipp = hippocampus; NAcc = nucleus accumbens; Put = Putamen; SN = substantia nigra; VP = ventral pallidum; VTA = ventral tegmental area

Figure 2

Schematic illustration of dopamine synapse on striatal medium spiny neuron. DA stimulation of D1-like receptors increases the activity of adenylate cyclase, while stimulation of D2-like receptors suppresses adenylate cyclase activity. DA may be removed from the synapse either by reuptake via the DA transporter or degradation by monoamine oxidase, resulting in the DA metabolite of homovanillic acid. Psychostimulants increase synaptic DA by blocking DAT function, while monaomine oxidase inhibitors block MAO activity and pramipexole inhibits DA autoreceptors. AC = adenylate cyclase; DAT = DA transporter; DOPA = 3,4-dihydroxyphenylalanine; HVA = homovanillic acid; MAO = monoamine oxidase; MAOI = monoamine oxidase inhibitor; MSN = medium spiny neuron; TH = tyrosine hydroxlase

Figure 3

Schematic diagram of effort-based decision-making paradigms. Animals may choose between a smaller food reward that is readily available (LC/LR option) or a greater food reward that can only be obtained after climbing over a barrier (HC/HR option). Control rats choose the HC/HR option approximately 90% of the time, while DA depleted rats show a strong preference for the LC/LR option.

Figure 4

Neural network model of normative effort-based decision-making. (A) ACC and amygdala estimate of costs and benefits of potential options. This information is relayed to NAcc, where it may be modulated by NAcc DA. Upon excitation, NAcc sends GABAergic signals that relay through VP to VTA, resulting in burst firing. (B) Theoretical model of how mesolimbic DA may help organisms overcome response costs adapted from (Phillips et al., 2007). The solid black line represents the cost/benefit calculation determined by ACC under normal DA levels. The dotted gray line shows how increases in striatal DA results in an increase in perceived net-value, despite increasing response costs. Conversely, the solid gray line shows how this value is significantly diminished under conditions of reduced DA striatal DA, particularly as response costs increase. ACC = anterior cingulate cortex; Amyg = amygdala; DA = dopamine projections; GABA = GABAergic projections; Glu = glutamatergic projections; NAcc = nucleus accumbens; VP = ventral pallidum; VTA = ventral tegmental area.