Dear reader of ADxS.org, please excuse the disruption.
ADxS.org needs about $63500 in 2024. In 2023 we received donations of about $ 32200. Unfortunately, 99.8% of our readers do not donate. If everyone who reads this request makes a small contribution, our fundraising campaign for 2024 would be over after a few days. This donation request is displayed 23,000 times a week, but only 75 people donate. If you find ADxS.org useful, please take a minute and support ADxS.org with your donation. Thank you!
Since 01.06.2021 ADxS.org is supported by the non-profit ADxS e.V..
Tonic firing refers to the sustained activity of a dopamine neuron at 0.2-10 Hz, which is mediated by cell-autonomous pacemaker impulses independent of stimuli.
Burst firing is characterized by short bursts of action potentials (3 - 10 spikes, >10 Hz) of a dopamine neuron in response to environmental stimuli. Bursts are usually caused by the activation of NMDA receptors via excitatory inputs. Burst firing is sometimes referred to as phasic firing, which emphasizes the synchrony of activity of dopamine neurons due to common inputs.1
Tonic and phasic firing must be distinguished from tonic and phasic release. Firing rates of somatic dopamine neurons are not linearly converted into axonal dopamine release, as the release is subject to a strong short-term depression. In addition, almost a third of tonic release occurs without somatic firing.1
Phasic dopamine is released into the synaptic cleft. If it is not reabsorbed there, e.g. in DAT-KO rats, it diffuses very slowly into the extracellular space and leads indirectly to extracellular dopamine. Tonic dopamine, on the other hand, is released directly into the extracellular space. Tonic dopamine therefore leads directly to extracellular dopamine.
Extracellular dopamine can regulate the phasic dopamine release of the transmitting neuron via autoreceptors of the presynapse. This influence is not limited to the immediate presynapse, but can also control neighboring neurons.
Sometimes extracellular dopamine is inaccurately referred to as tonic dopamine and vice versa. However, extracellular dopamine can also originate from other sources, e.g. from a DAT efflux or from diffusion of (previously phasically released) dopamine from the synaptic cleft.
In ADHD sufferers, a PET study found reduced dopamine release in the caudate nucleus at rest (“tonic dopamine”) and increased dopamine release during a flanker task (“phasic dopamine”). This tended to be similar in other parts of the striatum, but was not significant. This supports the hypothesis of overactive DAT.2
For more on tonic and phasic dopamine in various explanatory models of ADHD, see ADHD - Disorders of the dopamine system
Extracellular dopamine is dopamine located outside the cell. Complete removal of extracellular dopamine leads to Parkinson’s-like conditions, which suggests a functional role for extracellular (= basal) dopamine3
There is disagreement about the origin of extracellular dopamine.
The different results could be due to the fact that low-frequency electrical stimulation is often used in laboratory studies to mimic tonic release. However, this does not mimic the stochastic feature of random activation of neurons, which is typical for tonic release, but rather many axons are recruited simultaneously (due to the simultaneous electrical signal), thus mimicking the essential feature of a phasic release. In addition, frequencies are used that are higher than those normally used in vivo for tonic release.
Furthermore, there is a very different concept or definition of tonic and phasic dopamine (bursts), which makes it very difficult to find uniform results (see Marinelli below in the paragraph Bursts are dimensional.
Conception: extracellular dopamine results from tonic firing
In vivo (in living animals, authoritative), one study found that inactivation of the ventral pallidum (which only increases population activity, i.e. the amount of neurons ready to fire and subsequently tonically active, but not the bursts themselves) caused a significant increase in extracellular dopamine levels in the nucleus accumbens of up to around 50%, which persisted throughout the 1-hour recording period. In contrast, provocation of phasic dopamine release did not increase extracellular dopamine levels.4 According to this model, dopamine released phasically in vivo is reabsorbed by DAT before it leaves the synapse. This view was shared by voices according to which, in addition to the tonic release of dopamine (in rather small amounts) via varicosities directly into the extracellular space, a large amount of dopamine is released into the synaptic cleft during a (phasic) dopamine burst, from where it diffuses into the extracellular space.5
Extracellular dopamine is therefore the result of a tonic and/or phasic release of dopamine.
This is in line with the view of Gatzke-Kopp et al. and Liu et al.6 According to this view, the basal dopamine level is made up of a large number of small, short-lived tonic dopamine spikes.
However, the tonic signaling is mediated by these short-lived dopamine signals near the release sites, not by the extracellular dopamine level itself1
Conception: extracellular dopamine results from phasic firing
Another study on the origin of extracellular dopamine found that in vivo it largely results from phasic dopamine releases averaged over time, rather than from tonic release by single spikes.3 Marinelli et al7 consider it unlikely that individual (tonic) spike activity leads to significant dopamine release, partly because of the low presynaptic release probability and the tight control of dopamine diffusion by DAT.
Even in vitro (in the laboratory, less significant), phasic dopamine provoked by electrical stimulation is said to contribute more to extracellular dopamine than tonic dopamine.8 The basal (extracellular) dopamine level is therefore generated:
70% through the firing of action potentials (bursts)
Blocking the action potentials reduced the extracellular dopamine level by 70 %9
up to 30 % independent of action potentials and the proteins of the active zone RIM and Munc, for example through spontaneous vesicular fusion (tonic release)
According to Floresco et al., even high tonic firing is not sufficient for extracellular dopamine levels to be high enough to activate the D2 autoreceptors. Other influences, such as DAT efflux, may be necessary for this. An increase in population activity, which increases tonic firing, also increases dopamine efflux in the ventral striatum. This is consistent with the fact that tonic extrasynaptic dopamine levels are little affected by DAT-DA reuptake4
Contrary to this, Sulzer et al. describe dopamine reuptake in the intact striatum as the primary clearance mechanism of tonically released dopamine.10
Conception: extracellular dopamine results from tonic and phasic firing, efflux and volume transmission
We assume that extracellular dopamine has a variety of dopamine sources:
tonic firing, albeit possibly on a rather small scale
phasic firing
Efflux from dopamine transporters, possibly also from noradrenaline transporters
Volume transmission
Dopamine receptors can also be located outside the synaptic cleft and take up dopamine from distant sources (volume transmission). The prerequisite for volume transmission is said to be “open” synapses that allow spillover of the emitted dopamine into the extracellular space. These are frequent in the dopamine system.11
Estimates of extracellular dopamine levels in the striatum are consolidated in the low nanomolar range (20-30 nM)3. Individual data ranged from 4.2 nM12, 2 to 20 nM1, 5 to 26 nM1314 or up to 2.5 μM3
One study measured tonic dopamine concentrations of 90 nM (± 9 nM) and a dopamine diffusion coefficient of 1.05 (± 0.09 ×10-6 cm²/s) in the mouse brain in vivo using “fast-scan controlled absorption voltammetry”.15
The extracellular DA concentration is lower in the NAc than in the dorsal striatum.8
Here, too, there are very different opinions. In addition, many studies and publications do not distinguish linguistically between tonic dopamine firing and extracellular dopamine levels, so it is not always clear what is actually meant.
We believe that most of the studies describing the effect of tonic dopamine, without defining “tonic” more precisely, rather refer to the effect of extracellular dopamine levels. We therefore reproduce these studies here. Where studies explicitly mention tonicfiring, we have made this clear. Otherwise, we have put “tonic” in quotation marks.
According to one view, tonic signaling is not mediated by extracellular dopamine levels, but by the short-lived tonic dopamine signals themselves, near the release sites1
According to another view, the extracellular dopamine level downregulates the phasic dopamine responses triggered by stimuli via D2 autoreceptors. “Tonic” dopamine (“tonic” means: unclear whether the tonic signal itself or a resulting extracellular dopamine level) mediates the regulatory (inhibitory) control of the PFC on the ventral striatum, i.e. inhibits the (phasic) activity of the striatum. In response to reward stimuli, the striatum fires phasically in a dopaminergic manner and activates dopaminergic postsynaptic receptors. The “tonic” control is therefore inhibitory and modulates the excitatory phasic firing in response to reward stimuli6
In the NAc envelope, reduced tonic DA release correlated with increased DA release during phasic stimulation.
In the dorsolateral striatum, which exhibited higher tonic DA release, burst-evoked DA release was relatively constant over a wide frequency range (5-80 Hz) and burst lengths (1-20 spikes/burst). This is consistent with previous studies showing an inverse relationship between the probability of release elicited by action potentials and the degree of frequency-dependent release.
Blockade of DAT or D2 autoreceptors primarily enhanced tonic dopamine firing. Blockade of nicotinic β2-acetylcholine receptors suppressed tonic dopamine firing. Suppression of tonic dopamine release increased the contrast between phasic and tonic dopamine firing.8
Activation of glutamate receptors in the VTA stimulates tonic (basal) dopamine release in terminal regions, including the nucleus accumbens. Hindbrain inputs from the laterodorsal tegmental nucleus (LDT) regulate the triggering of phasic dopamine activity in the VTA. Administration of iGluR agonists (AMPA or NMDA) increased tonic dopamine in the NAc and decreased LDT-mediated phasic dopamine in the NAc. D2 autoreceptor agonist (quinpirole) administration into the VTA following NMDA administration limited the NMDA-mediated increase in tonic dopamine in the NAc and partially reversed the NMDA-induced attenuation of LDT-mediated phasic NAc dopamine. Infusion of the D2 autoreceptor agonist quinpirole alone attenuated tonic and LDT-mediated phasic dopamine.16
An abnormally low tonic extracellular dopamine level (e.g. in ADHD) leads to an upregulation of the D2 autoreceptors, so that stimulation-induced phasic dopamine is increased.17
According to Grace, ADHD is characterized by this pattern of abnormally low tonic (extracellular) dopamine levels triggering excessive phasic dopamine release.18 However, we suspect that this is only one of several options for a dopaminergic imbalance that triggers ADHD. It is possible that subtypes show different patterns here.
6.1.3.2. Extracellular (“tonic”) dopamine regulates the use of what has been learned / rigidity¶
Changing a response strategy due to changes in the criteria for achieving goals requires a reduction in “tonic” dopamine.19 Permanently increased “tonic” dopamine therefore causes rigidity.20 Previously learned associations between cue and reward were maintained in vivo by continuous tonic firing from the VTA into the PFC, even if they are no longer valid.21
One study compared DAT-KD mice and wild-type mice in an environment where food could only be obtained by pressing two levers. The number of lever presses required to obtain food alternated frequently between the two levers. DAT-KD mice (= reduced DA reuptake = increased extracellular dopamine) pressed the expensive levers more frequently and thus worked harder for their food. The speed of response to changes in lever cost was comparably fast. DAT-KD mice showed normal learning from recent reward history: lever choice was less strongly linked to reward history. Increased extracellular dopamine appears to reduce the ability to benefit from what is learned.22
We believe this further suggests that phasic dopamine is not impaired in the DAT-KD mouse, as the hypothesis that dopamine plays a role in action learning is largely based on recordings of phasic dopamine responses.2324
6.1.3.3. Extracellular (“tonic”) dopamine encodes measure of motivation (?)¶
According to Berke, the hypothesis that extracellular dopamine encodes the level of motivation is contradicted by the fact that extracellular dopamine levels hardly change within an individual and dopaminergic cells cannot switch between active and quiescent states. Thus, no extracellular level changes have been observed due to the interaction of several dopaminergic tonic sources in the sense of a cluster.25 The latter view, however, is in clear contrast to the descriptions of Grace and Marinelli, among others.
6.1.3.4. Extracellular (“tonic”) dopamine encodes time perception¶
Dopamine influences interval timing. Similar to reinforcement learning, extracellular dopamine also appears to act in two directions here:26
Dopamine modulates the speed of the internal timing mechanism
acute increase in DA accelerates the internal clock
acute DA drop slows down the internal clock
Parkinson’s patients without medication (chronically DA-depleted) show an altered perception of time: the “central tendency” is more pronounced, which means that when learning intervals of different durations, shorter intervals are overproduced and the longer intervals are underproduced. DA supplementation in these patients corrects this change in timing.27
Low to moderate levels of stress increase extracellular7dopamine levels in the nucleus accumbens2829 , but only in the NAc shell, not in the NAc nucleus3031 and PFC2829 , while high levels of stress (intense, chronic or unpredictable) decrease dopamine levels3233 . The increase in dopamine levels is greater in the PFC than in the striatum; within the striatal complex, it is greatest in the NAc shell.347
Most stressors increase extracellular dopamine through an increase in dopamine efflux, an increase in neuronal activity in total firing rate and/or bursts:7
Food restriction/withdrawal
Bondage stress
Social defeats
Cold swimming Chronic stressors (chronic cold exposure, chronic mild stress) have been shown to decrease population activity, i.e. the number of active neurons, but only in the medial and central VTA, not in the lateral VTA, and without decreasing the firing frequency. Bursts were slightly increased during chronic cold exposure.
Stressors that increase dopamine firing also increase the risk of addiction and addiction relapses, which are prevented by blocking dopamine receptors.
Tonic release generates short-lived dopamine transients of a few milliseconds at irregular intervals, in which only some of the dopamine neurons are involved. Tonic dopamine is not released into the synapse but into the extracellular space, where it is rapidly distributed. The basal (extracellular) dopamine level is the result of a balance between tonic release and DAT (and NET) reuptake. The basal dopamine level is below the activation threshold of most dopamine receptors. Presumably, the basal dopamine level is composed of a large number of small, short-lived dopamine spikes. Tonic signaling is likely to be mediated by these short-lived dopamine signals near the release sites, not by the basal dopamine level itself1
6.2.1. Source and destination of the tonic firing¶
Tonic dopamine release occurs in particular from varicosities, i.e. extrasynaptically, into the extracellular space. From there, dopamine diffuses to autoreceptors or to (extrasynaptic) receptors of its own neuron or other, sometimes relatively distant, neurons (volume transmission). Dopamine is degraded in the extracellular space by COMT35
There are controversial hypotheses about the purpose and effect of tonic firing:
As tonic dopamine is not released into the synapse, it does not trigger a signal at the postsynaptic receptors. It only activates presynaptic autoreceptors (also from neighboring nerve cells), which in turn can slow down the phasic dopamine release of your nerve cell (negative feedback).35
The tonic dopamine creates a constant dopamine level of 4 to 10 nanoMol in the downstream neuronal structures (e.g. in the nucleus accumbens), which is sufficient to activate high-affinity D2 receptors, but not the less affine D1 receptors.36 According to other
Tonic nanomolar dopamine acted at D1 receptors in crayfish to support the maintenance of phase in the lateral pyloric neuron in an activity-dependent manner. Phasic dopamine can apparently activate homeostatic mechanisms that maintain motor network performance.37
The phasic “background” dopamine level does not appear to trigger D2 autoreceptor feedback depression. Apparently, the extracellular level achieved by tonic firing is too low to stimulate the low-affinity D2 receptors38
In vivo, VTA dopamine neurons of anesthetized adult rats fire at a frequency of around 4.5 Hz (0.5 to 10 Hz). This activity is subject to a normal distribution. Most cells fire at around 4 Hz39 / 4 Hz 4041 / 4 to 5 Hz7 Liu et al call 0.2 to 10 Hz.9
This time interval of tonic firing (250 ms / 4 Hz) allows for maximal autoinhibition of tonic firing via D2 autoreceptors, as the suppression of DA release by D2 autoreceptors in vivo started after about 150 to 300 ms and lasted for about 600 ms. In contrast, bursts are usually completed before autoinhibition begins.38 In vivo, dopamine neurons fire tonically in a slow, irregular firing pattern due to local circuits and afferent GABAergic inputs. Without these influences, a very regular, slow pacemaker pattern exists in vitro.42
For comparison: The noradrenergic tonic frequency of the locus coeruleus is typically between 1 and 6 Hz and can rise to 8 to 10 Hz under considerable stress.43
Around half4239 to 98 %1 of the dopaminergic VTA neurons are active and fire spontaneously. The inactive neurons do not fire spontaneously4239 because they are constantly hyperpolarized by an inhibitory GABAergic influence from the ventral pallidum and thus kept inactive. Activation of GABA-B receptors inhibits tonic and phasic dopamine release.38 If these afferents from the pallidum are suppressed, the neurons are freed from GABAergic inhibition and fire spontaneously again.44445
Changes in tonic dopamine efflux occur on a much slower time scale than changes in phasic dopamine efflux.4
Tonic dopamine in the nucleus accumbens is likely to be regulated by glutamatergic afferents from the PFC46 Tonic dopamine firing in the midbrain continuously tracks reward levels that change from moment to moment. Tonic activity is more strongly evoked by spontaneous non-burst spikes than by burst spikes that produce conventional phasic activity.47
D-AMP and MPH injected into the abdominal cavity influence tonic firing.48 Since the study was performed in living, immobilized and partially anesthetized rodents, we do not assume a phasic release.
D-AMP:
2.5 mg/kg:
reduced tonic firing in the neostriatum
reduced tonic firing in substantia nigra
5 mg/kg and 7.5 mg/kg:
increased tonic firing in the neostriatum
reduced tonic firing in substantia nigra MPH:
10 mg/kg:
reduced tonic firing in the neostriatum
reduced tonic firing in substantia nigra
20 mg/kg and 25 mg/kg:
increased tonic firing in the neostriatum
reduced tonic firing in substantia nigra
It remains to be seen whether this study is transferable to AMP in drug form.
In studies on the effects of amphetamine, it must always be borne in mind that these
AMP is usually used in significantly higher doses than for ADHD medication
use unretarded / not prolonged-acting AMP via prodrug
Inject AMP often, which again results in much faster metabolization
these 3 factors multiply in their effect
There is no doubt that AMP in drug form has a different effect than AMP in drug form.
6.2.5. Behavioral effect of tonic firing / extracellular dopamine levels¶
Due to the ambiguity of what most studies mean by “tonic” dopamine when describing its effects, we have summarized the results above under effects of extracellular dopamine, unless they specifically mention tonic firing.
Tonically released dopamine appears to act as a neuromodulator to modulate the excitation of striatal projection neurons, which is probably required for the correct selection and execution of voluntary movements.11
Tonic, but not phasic, stimulation of DA neurons in the VTA appears to regulate alcohol craving.49
High tonic dopaminergic activity of the VTA in vivo caused:43
a consumption of dopamine, which however regenerated faster than noradrenaline
reduced phasic dopaminergic transmission during subsequent ignition events
High tonic noradrenergic activity of the locus coeruleus in vivo caused:43
depletion of the noradrenaline reserves
a desensitization of the noradrenergic postsynaptic excitability response
significantly reduced phasic noradrenergic transmission during subsequent ignition events Acute stress primarily affected the postsynaptic response and reduced phasic NE release43
In vivo, VTA dopamine neurons of anesthetized adult rats fire 2 to 6 burst spikes at 15 Hz (10 to 20 Hz).397 Other sources speak of 3 to 10 action potentials at 12 to 25 Hz42 or around 20 Hz or more40, followed by prolonged post-burst inhibition.42
During burst firing, a dopamine release of up to 0.2-1.0 μM with a half-life of 0.25 ms was measured50 and an extracellular dopamine level of 8-15 μM.13 Other sources report up to 150 nM extracellular dopamine in the nucleus accumbens with (stronger) in vitro stimulation of the VTA with NMDA.14
This high dopamine input suppresses the subsequent release of dopamine for several seconds via D2 autoreceptors at dopaminergic terminals51, most likely by suppressing axonal Ca2+ currents.
6.3.4. Difference between phasic dopamine firing and bursts¶
Phasic release on the one hand and bursts (phasic firing) on the other correlate with each other, but must be distinguished1
Phasic release is dependent on the simultaneous recruitment of a tonically firing population of dopamine neurons and relies on synchrony between dopamine neurons. Phasic signaling does not require (repeated) burst firing of individual neurons.
Burst firing is the result of synchronized activation of individual dopamine releases from a large number of different neurons. As a rule, burst firing (by means of action potentials) is synchronized across all dopamine neurons. The first spike efficiently increases dopamine levels, while subsequent activity releases less dopamine due to less synchrony and the presence of refractory sites.
In the striatum, subsequent burst spikes increase the dopamine level in the striatum only slightly further, but serve to maintain the increased levels caused by the first spike, i.e. they prolong the dopamine’s retention time.
In the nucleus accumbens, however, the dopamine level continues to rise due to further burst spikes.
Phasic dopamine release (bursts) occurs from the vesicles into the synapse. Stimuli such as reward or other stimuli activate short bursts of action potentials from dopaminergic neurons. These dopamine bursts occur at around 20 Hz or more8, last less than 200 ms and release large amounts of dopamine from storage vesicles in the presynapse into the synaptic cleft. This phasically released dopamine crosses the synaptic cleft and activates receptors at the postsynapse. After release by the receptors, the dopamine is taken up from the synaptic cleft back into the presynapse by dopamine transporters (reuptake). In smaller quantities, it diffuses out of the synapse into the extracellular space or is degraded (albeit subordinately) by COMT located in the synaptic cleft.3552
The number of bursts correlates only roughly with the total firing rate (correlation: r = 0.38 to 0.41). The total firing rate correlates best with the non-burst activity (correlation: 0.96).
A helpful illustration of the tonic and phasic firing of a dopamine neuron can be found in Marinelli et al.537
Dopamine release within a single neuron occurs robustly in response to an initial activation, but then quickly subsides for several tens of seconds. It follows that even tonic firing leads to a seconds-long exhaustion of the respective dopamine release site, and dopamine release in response to each action potential is largely determined by the recovery of these release sites. Neurons with lower (tonic) spontaneous activity therefore contribute more to phasic release because their release-ready vesicle pool is less depleted when the synchronizing stimulus arrives. In burst firing, only the first few action potentials lead to significant dopamine release from a single axon. Thus, it is the synchrony of group firing, rather than the firing pattern of individual neurons, that dominates signaling during phasic release. This view is supported by the fact that burst firing is severely impaired in mice lacking NMDA receptors, while phasic dopamine transients and the behaviors they mediate persist. Phasic release is the result of the simultaneous activation of a large number of dopamine release sites. The dopamine reuptake mechanisms are temporarily overridden. This leads to considerable crosstalk between the dopamine signaling areas and causes prolonged dopamine dwell times. In phasic signaling, the rapid increase in dopamine in spatial areas of several micrometers can lead to the activation of dopamine receptors that are somewhat distant from the release sites. A prerequisite for phasic release and signaling is synchrony of release across dopamine neuron populations.1
In the dorsolateral striatum, increasing the burst length from 1 to 10 pulses (at 20 Hz) only moderately enhanced the dopamine signal, whereas in the nucleus accumbens, dopamine release increased strongly with increasing burst length (even more strongly in the NAc shell than in the NAc nucleus).40
According to Marinelli et al. bursts differ from cell to cell, even under resting conditions, in a variety of parameters. In addition, bursting is dimensional, i.e. it can be measured in a continuum according to the quantity or extent of the various parameters, but not on the basis of exact criteria.7
There is a very gradual continuum between non-bursts and bursts. Bursting is neither an “all-or-nothing” phenomenon nor does it exhibit a bimodal distribution across the neuron population.
As there is no universally accepted definition of bursts, studies can produce seemingly contradictory or incompatible results due to the different terms used.
Naturally, the frequency of burst events and their characteristics change due to stimuli or environmental conditions - this change is the very definition of bursts.
Nevertheless, it is often described how cells that are in a “non-burst mode” (not emitting spikes in clusters) suddenly change their firing pattern to a “burst mode”.
Very high or low extracellular dopamine controls phasic dopamine via D2 autoreceptors. See above.
The phasic pattern is dependent on glutamatergic afferent inputs39, especially from the pedunculopontine tegmentum.54
Glutamatergic input from the pedunculopontine tegmentum acts to activate dopamine neurons at glutamatergic NMDA receptors to produce phasic firing bursts that generate the behaviorally relevant fast (phasic) dopamine response. GABA from the ventral pallidum inhibits the dopamine neurons of the VTA by placing subsets of dopamine neurons in a hyperpolarized, non-firing (silent) state through magnesium blockade of the NMDA channel so that they cannot participate in the phasic dopaminergic response. In a safe environment, the inhibitory GABAergic input is high and the number of dopamine neurons ready to fire phasically is therefore small. Conspicuous stimuli then only trigger a calm orientation response. In a threatening environment, conspicuous stimuli trigger a phasic reaction more easily, as many dopamine neurons are ready to fire. Alertness to the environment is increased and the organism can adapt more quickly to the stimulus and prepare an appropriate response.42
Synaptotagmin-7 appears to co-control phasic dopamine firing.55
In vivo, ATP-sensitive potassium channels (KATP channels) in DA neurons of the medial substantia nigra projecting to the dorsomedial striatum control switching from tonic to phasic activity. Coactivation of ATP-sensitive potassium channels played an essential role in NMDA receptor-induced bursts in vitro.56
In Parkinson’s disease, transcriptional dysregulation of ATP-sensitive potassium channel subunits and a high proportion of burst discharge patterns were found in vivo in surviving human nigrostriatal DA neurons. In vulnerable substantia nigra DA neurons, a selective activation of ATP-sensitive potassium channels by oxidative stress was found. These results could indicate a possible role of ATP-sensitive potassium channels in the pathomechanism of Parkinson’s disease either due to burst activity-induced hyperexcitability (‘stressful bursting’ ) or through a compensatory, homeostatic adaptation of the DA system in the disease process56
Ghrelin in the lateral hypothalamus (but not in the VTA) increased the strength of food-induced dopamine spikes, LV ghrelin receptor antagonists reduced them. Intra-VTA orexin-A also increased them, intra-VTA blockade of orexin receptors attenuated them. Food restriction increased the strength of food-induced dopamine spikes.57
6.3.7.1. Behavior regulation through phasic dopamine¶
Phasic dopamine signals are relevant for synaptic plasticity, reward processing and behavioral learning.5859Phasic stimulation (continuous as well as after decisions that were not previously associated with a reward) facilitates decisions that deviate from previously learned associations (enables new experiences).21
Larger phasic responses of dopamine cells to triggers predict shorter reaction times for the same task.60
Phasic dopamine encodes the quantitative reward prediction error (RPE).616263 The RPE is the difference between the reward received and the predicted reward64 Dopamine neurons in the midbrain, as well as a subpopulation of dopamine neurons in the striatum, amygdala and PFC
Fire dopaminergically when a reward is higher than expected
Remain unchanged if the reward corresponds to the expectation
Reduce their dopaminergic activity when the reward is less than expected
The most probable rewards form mean values. If there are two alternative expected reward amounts65
No reaction if one of the two probable rewards or exactly the middle value between the rewards occurs
A positive reaction for rewards that are slightly higher than the lower reward (i.e. more than if the upper reward occurred, even though the reward is lower)
A negative reaction for rewards that are slightly below the upper reward (i.e. less than if the lower reward occurred, even though the reward is higher)
A deviation that is close to an expected result is thus considered relative to this reference value, so that it is updated, while a result that is equally far away from both expected reference values has no novelty value in relation to the reference values and therefore does not trigger an update
During the RPE, dopamine neurons respond with short, phasic bursts of activity, e.g. to appetizing stimuli. If a previously irrelevant stimulus has been learned to be associated with a reward, the dopamine neurons shift their phasic activation from the time of reward receipt to the time of presentation of this predictive cue stimulus. The unexpected absence of the reward leads to a suppression of the activity of the dopamine neurons.64
When reward onset is uncertain and the probability of onset decreases with time (as in people who give up waiting for an unreliable bus), the longer the waiting time, the greater the prediction error and dopamine response at reward onset.66
Reward-predictive responses in the amygdala also correlate with the temporal probability of reward67
Delayed reward also showed a reduction in phasic dopamine signaling in the NAc. However, the dopamine difference was only half as large as in the prediction of low or high cost of reward attainment68
In addition to the quantitative reward prediction error (RPE), dopamine also appears to mediate a prediction error for sensory prediction errors.697071
6.3.7.1.4. Phasic dopamine encodes avoidance value of aversive stimuli¶
Phasic dopamine probably encodes the avoidance of an aversive stimulus rather than aversive stimuli themselves.7261 or the physical / sensory intensity of a punishment73
Dopaminergic firing to aversive stimuli did not encode the level of aversiveness
Aversive stimuli cause dopamine release in only a few dopamine neurons
Part of the dopaminergic cells of the substantia nigra pars compacta dorsolateral fired in response to aversive, surprising or alarming events.7475 Another part of the dopaminergic cells, especially ventromedially and in the VTA, stopped firing in response to aversive stimuli.
One study reports on different dopaminergic neurons, whereby
that excites you through rewards and inhibits you through aversive stimuli and
the others were activated by both types of stimuli.
The dopamine neurons that were excited by aversive stimuli or stimuli predicting aversive events were more likely to be found in the dorsolateral substantia nigra pars compacta, while neurons that were inhibited by these stimuli were more likely to be found ventromedially, including in the ventromedial VTA.76
6.3.7.1.5. Phasic dopamine controls the learning of a response strategy to reinforcement¶
Learning a response strategy to reinforcement
Takes place via phasic dopamine in the nucleus accumbens via D1 receptors19
A large proportion of the dopaminergic neurons of the substantia nigra pars compacta, which do not overlap with the dopamine neurons responding to reward (expectation), fire phasically prior to the initiation of a self-directed movement. This activity was not action-specific and represented the strength of future movements. Inhibition of these cells decreased the probability and strength of future movements, while brief activation increased the probability and strength of future movements.77 This may be the key to the movement impairment observed in Parkinson’s disease.
6.3.7.1.7. Phasic (mesolimbic) DA encodes the value of work¶
One view proposes that mesolimbic phasic dopamine encodes the value of work required to achieve a goal, i.e. the need to invest time and effort to obtain the reward.78Dopamine levels only increase with signals that prompt exercise, but not with signals that prompt rest, even if these indicate a similar future reward.79
6.3.7.1.8. Dopamine response to neutral stimuli: Novelty¶
Dopamine neurons also react to “neutral” stimuli without positive or negative valence, e.g:7
Novelty value / unfamiliarity
Alarm signals
This makes it possible to learn the value of new stimuli or changes in behavior.80
When previously neutral stimuli reliably predict a rewarding or aversive stimulus, it turns them into conditioned predictors that trigger a firing response similar to the predicted stimulus.81
These reactions are presumably controlled by the cortex via the superior colliculus.82
6.3.7.1.9. Dopamine encodes the benefit of an expenditure of resources¶
Berke25 sheds light on the hypothesis that dopamine encodes the benefit of consuming a limited resource, namely
economic (distribution of resources) and
motivational (whether it is worth spending resources)83
The hypothesis by Beeler et al that obesity could be caused by a lack of motivation to exercise, whereby a lack of dopamine correlates with a lack of motivation, must be countered by the fact that ADHD, which is also associated with a lack of dopamine, is associated with the symptom of “always having to be active” and hyperactivity. Dopamine deficiency is therefore not consistently causal for a “sedentary lifestyle that inhibits energy consumption”
The circuits within the striatum are organized hierarchically: The ventral striatum influences dopamine cells, which in turn project into the dorsal striatum. In primates, the ventromedial striatum consists of the shell, which receives limited input from the cortex, midbrain and thalamus, and the core. The cortex dopaminergically influences the nucleus, the nucleus influences the central striatum, and the central striatum influences the dorsolateral striatum.8485 This means that the decision to start work can also have the effect of reinforcing the specific, shorter movements required. Overall, however, dopamine provides “activating” signals (which increase the likelihood that a decision will be made) rather than “directional” signals that indicate how resources should be used.
6.3.7.1.9.1. Dorsolateral striatum: dopamine encodes the resource movement¶
In the dorsolateral striatum, dopamine encodes the resource of movement, which is limited due to energy expenditure and the incompatibility of several actions at the same time.8687 An increase in dopamine increases the likelihood that an individual will consider the energy expenditure for a movement to be worthwhile.8887838990 If higher dopamine codes for “exercise is worthwhile”, there is also a correlation between dopamine and the exercise itself, but this is not directly causal.
Slower dopamine activity in the substantia nigra (which projects to the dorsolateral striatum) appears to reflect general behavioral activation.9192
Subpopulations of dopamine neurons are heterogeneously activated or inhibited (subsecond to second range) during different task events involving movements that engage a large number of muscles (e.g. 35 or more) and sensory receptors.
However, dopamine neurons are not activated during movement sequences of well-controlled arm and eye movements with few muscles involved.
The substantia nigra neurons showed different impulse rates:
pars reticulata: pulse duration 0.6 to 1.0 ms; pulse rate 68/s (23/s to 145/s)
pars compacta: pulse duration longer; pulse rate below 8/s
6.3.7.1.9.2. Dorsomedial striatum: dopamine encodes the resource of cognitive processes¶
According to Berke, dopamine in the dorsomedial striatum encodes the resources of cognitive processes such as attention (which is limited by definition)93 and working memory.94Dopamine codes the attention to conspicuous external cues. The conscious activation of cognitive control processes is complex.95Dopamine codes - particularly in the dorsomedial striatum96 - that it is worth making this effort, for example whether the effort of cognitively demanding, model-based decision-making strategies is worthwhile. Dopamine deficiency, on the other hand, causes such cues, which normally trigger orientation movements, to be neglected - as if they deserved less attention97
One study suggests that the dorsal ACC (dACC) should increase its proportion of
Reward processing
Performance monitoring
cognitive control
Choice of action
the evaluation of the expected value of control (EVC) is performed solely on the basis of a single value. The presented normative model of EVC integrates three critical factors:
the expected profit from a controlled process
the amount of control that must be invested to achieve this profit
the costs in the form of cognitive effort.
The ACC is primarily controlled by glumaterg and GABAerg.
6.3.7.1.9.3. Ventral striatum (nucleus accumbens): Dopamine encodes the resource time and reward / motivation¶
According to Berke, dopamine encodes the resource time in the nucleus accumbens. Some rewards require long preparatory work for actions that are not rewarded in detail, e.g. foraging. A decision in favor of such time-consuming work means foregoing other advantageous ways of spending time. A high mesolimbic dopamine level encodes that it is worthwhile to do time-consuming, strenuous work for a distant goal. If the mesolimbic dopamine level drops, the interest in long-term rewards decreases.
Phasic dopamine in the ventral striatum (nucleus accumbens) encodes reward or motivation.87
Simple actions with quick rewards do not require mesolimbic dopamine.98
Dopamine ramps are not inconsistent with the reward prediction error hypothesis. RPEs can increase over the course of a trial due to sensory feedback. A gradual attenuation of sensory feedback leads to a DA bump.99
6.3.7.1.11. Lack of phasic dopamine with increased tonic dopamine correlates with psychosis symptoms¶
A lack of phasic dopamine so severe that bursts are no longer possible, combined with high tonic firing rates, can reduce the dynamic range of DA signaling and ultimately contribute to psychosis-like behaviors.
Such a condition can arise in chronic NMDA receptor deficiency through presynaptic adaptation reactions of dopaminergic neurons.100
6.3.7.2. Control of neurophysiological factors by phasic dopamine¶
A computer model found that phasic dopamine firing decreased average D2 receptor occupancy and increased D1 receptor occupancy compared to equivalent tonic firing. Receptor occupancy depended critically on synchrony and balance between tonic and phasic firing modes.101
Phasic (non-tonic) activity of dopamine neurons induces the formation of mesofrontal axonal boutons in juvenile rodents. In adults, the effect of phasic activity decreases. Inhibition of D2 dopamine receptors restores this plasticity in adulthood102
One study suggests that these heightened phasic responses lead to hypersensitivity to environmental stimuli in ADHD. Stimuli that cause moderate arousal of the brain lead to well-functioning performance, while either too little or too much stimuli weaken cognitive performance. Strong, salient stimuli can easily disrupt attention, while a low-stimulus environment causes low arousal, which is usually compensated by hyperactivity. Stochastic resonance is the phenomenon that causes moderate noise to facilitate stimulus discrimination and cognitive performance. Computational modeling shows that more noise is required for stochastic resonance to occur in dopamine-deficient neural systems in ADHD. This prediction is supported by empirical data17
An external, sufficiently strong event (punishments, rewards and novel stimuli) triggers several responses of dopaminergic neurons.
First a short, unspecific, prediction-dependent arousal.
For primary rewards and reward-predicting stimuli, this is followed by the biphasic reward prediction error (RPE) signal. The RPE response lasts less than 1 second.
These two initial reactions are followed by slower excitations or inhibitions with a large number of sufficiently strong events and processes, including sensory stimuli and movements. These reflect a general function of arousal and behavioral activation.
Aversive or high-intensity stimuli evoked a three-phase sequence of activation-suppression-activation over a period of 40 to 700 ms:
* Start phase: Activation with short latencies (40-120 ms)
* Encodes the sensory intensity
* Middle phase: (between 150 and 250 ms)
* Codes the motivational value
* Activation with appetitive (pleasant) stimuli
* Suppression of aversive (unpleasant) and neutral stimuli
* Reward prediction error103
* Activity increased for 100 to 200 ms if reward or reward prediction stimulus is better than predicted
* Activity unchanged if events have the same reward value as predicted
* Activity briefly dampened when events have lower reward value than predicted
* Late phase:
* Moderate “rebound” after strong suppression
* Strong activation through high reward is often followed by suppression
A simulation of bursts in the VTA caused long-lasting (up to one hour) elevated extracellular dopamine levels in the nucleus accumbens and PFC after the first dopamine peak, presumably due to reduced dopamine reuptake. Tetrodotoxin (which prevents action potentials by blocking sodium channels) terminated the increase in dopamine levels.104
In the NAc, dopamine levels of 10 μM - 1-10 μM have been reported as a result of bursts105 - cause reduced DAT expression in the midbrain in vitro. DAT inactivation by dopamine was mediated by increased Rho-GTPase. Inactivation of Rho-GTPase suppressed dopamine-mediated DAT internalization. 104 Dopamine causes increased adenylyl cyclase activity via D1/D5 receptors, which in turn activates PKA.106 PKA deactivates Rho-GTPase through phosphorylation107
However, this was a fairly intense stimulation over 20 minutes each with a series of electrical pulses (pulse width = 1 ms, burst width = 200 ms, interburst interval (IBI) = 500 ms) or optogenetic pulses (a) 20 pulses (pulse width = 1 ms) at 100 Hz for 200 ms, IBI = 500 ms; and (b) 100 pulses (pulse width = 5 ms) at 20 Hz for 5 s, IBI = 10 s. This is considerably higher than the typical bursts in rats reported by Grace or Zhang (see above).
It is possible that this prolonged extracellular dopamine increase is required for the consolidation of learning experiences. This is because D1 and D2 dopamine antagonists administered systemically or directly into PFC or striatum (minutes) after behavioral training interfere with memory consolidation, whereas subcutaneous D2 antagonist administration enhances it.108109110Dopamine administration into the basolateral amygdala or NAc shell (but not into the NAc nucleus) after training improved memory performance, while non-selective dopamine receptor antagonists prevented this improvement. Apparently, simultaneous DA receptor activation in the NAc shell and the BLA is required for memory consolidation.111
This provides an interesting context to the common learning problems in ADHD, which is associated with decreased tonic/extracellular dopamine, and the cases of memory artists in ASD (movie cliché: Rainman), which is associated with increased extracellular dopamine
6.3.9. Speed of dopamine level change in the brain encodes different behaviors.¶
The speed at which dopamine levels change in the brain encodes different behaviors.
Encoding changes in dopamine levels
In the 10-minute range: the strength of motivation and behavioral activation
In the seconds range: the value of a future reward
In the sub-second range: the search for the reward
In the mPFC, the amplitude and direction of synaptic plasticity is determined by long-term (ten minutes) changes in dopamine levels, not by short-term changes.112113 In the striatum, on the other hand, a short DA pulse can cause a change in the direction of plasticity simultaneously with afferent stimulation.114
6.3.9.1. 10-Minute range coded Strength of motivation and behavioral activation¶
Encoding changes in dopamine levels
In the 10-minute range (ramps): Strength of motivation and behavioral activation
Activations are mediated by changes in dopamine levels in the nucleus accumbens (part of the basal ganglia in the striatum, part of the mesolimbic system) when the changes in dopamine levels occur at a (slow) rate in the 10-minute range. Slow dopamine level changes over time units of 10 minutes correlate with the reward rate, the strength of motivation and behavioral activity.115116
The cortical dopaminergic projection system appears to be geared to relatively slow changes in tonic DA concentration (over seconds to minutes) and is not sensitive to short (i.e., approximately 200 ms) phasic dopaminergic signals.117118 Consistent with this, the mPFC has a much lower DA clearance rate than the striatum at the same extracellular dopamine level119
The PFC dopamine appears to originate from a subpopulation of molecularly and functionally distinct DA neurons in the VTA that signal reward-related activity.120 This subpopulation can sustain tonic firing at high rates over prolonged periods and is not subject to D2 autoreceptor feedback control.
6.3.9.2. Seconds range encodes value of future reward¶
Phasic activations are mediated by changes in the dopamine level in the nucleus accumbens if the changes in the dopamine level occur at a (rapid) rate in the range of seconds. Rapid (relative) changes in dopamine levels every second mediate the evaluation of a future reward. The value of a future event is therefore estimated and encoded by changes in the value of dopamine in the range of seconds115
6.3.9.3. Subsecond range activated a. Search for reward and b. Movement¶
Even more short-term phasic changes in dopamine levels in the range of fractions of a second were observed in rats that were trained to respond to a signal in which they could request sugar or cocaine. The corresponding signal triggered an extremely rapid increase in dopamine levels in animals trained in this way (time ranges below one second). Only in animals trained in this way could a dopamine administration in the nucleus accumbens, which took place at the corresponding speed, trigger the search for the reward.121122
Specific other axons leading to the striatum respond to fast phasic dopamine signals to encode movement initiation.123
Differentiating the effect of dopamine according to the speed of the level increase on the one hand (seconds and sub-second range) and changes in the direction of the absolute level dimension on the other (10-minute time measure) can explain why dopamine is simultaneously relevant for (short-term) motivation and (long-term) learning.
6.3.10. Does acetylcholine switch between dopaminergic coding of reward prediction errors and learning?¶
Several reports show that dopamine encodes both the reward prediction error and value signals. It is possible that dopamine-receiving circuits can actively switch how they interpret dopamine. Evidence suggests that acetylcholine, among others, may have such a switching function.
While dopamine cells respond to unexpected signals with phasic spike bursts, cholinergic interneurons in the striatum show short pauses of about 150 ms during which they do not fire, and which do not scale with reward prediction error values.
These pauses of the cholinergic interneurons can be triggered by GABAergic neurons of the VTA as well as by “surprise” cells in the intralaminar thalamus. For example, GABA releasing neurons of the VTA projecting to the nucleus accumbens are able to inhibit cholinergic interneurons in the accumulation area to enhance learning of stimuli and outcomes. It is possible that these pauses act as association signals that promote learning. During pauses in cholinergic interneurons, the cessation of muscarinic blockade of synaptic plasticity appears to encode dopamine as a signal for learning. If the cholinergic interneurons fire, the release of dopamine terminals is controlled locally to influence ongoing behavioral performance. However, this is not yet certain.25124125126
The interaction of tonic and phasic dopamine in the striatum allows an optimal solution to the exploration/exploitation problem by estimating the true state of a noisy Gaussian diffusion process. In silicio (in the computer model), the association with phasic dopamine-dependent influences on corticostriatal plasticity showed performance at the level of the Kalman filter, which provides an optimal solution to the task.127
The decisive factor was a fluctuating level of tonic dopamine, which increased under unsafe conditions.
In the optimal range of tonic dopamine, exploration-exploitation decision making was mediated by the effects of tonic dopamine on the precision of the model action selection mechanism.
In the presence of uncertain reward, the reduced selectivity of the model allowed the disinhibition of several alternative actions that were explored at random.
When the reward was certain, the selectivity of the action selection circle was increased, which facilitated the utilization of the valuable choice.
6.5. ADHD explanation models according to tonic and phasic dopamine¶
In ADHD sufferers, a PET study found reduced dopamine release in the caudate nucleus at rest (“tonic dopamine”) and increased dopamine release during a flanker task (“phasic dopamine”). This tended to be similar in other parts of the striatum, but was not significant. This supports the hypothesis of overactive DAT.2 This supports the model of Grace et al.
Other models assume a dopamine deficit in tonic and phasic dopamine in ADHD.
6.6. For differentiation: tonic and phasic receptor types¶
The categorization of receptors into tonic and phasic receptors must be distinguished from phasic and tonic neurotransmitter release. The receptor categorization describes the reaction mode of receptors and has nothing to do with the release mode of neurotransmitters:
Marinelli M (2007): Dopaminergic reward pathways and the effects of stress. In: Al’Absi M, editor. Stress and Addiction: Biological and Psychological Mechanisms. Academic Press; Burlington: 2007. pp. 41–84 ↥
