Distinct prestimulus and poststimulus activation of VTA neurons correlates with stimulus detection

Abstract

Dopamine neurons of the ventral tegmental area (VTA) signal the occurrence of a reward-predicting conditioned stimulus (CS) with a subsecond duration increase in post-CS firing rate. Important theories about reward-prediction error and reward expectancy have been informed by the substantial number of studies that have examined post-CS phasic VTA neuron activity. On the other hand, the role of VTA neurons in anticipation of a reward-predicting CS and analysis of prestimulus spike rate rarely has been studied. We recorded from the VTA in rats during the 3-choice reaction time task, which has a fixed-duration prestimulus period and a difficult-to-detect stimulus. Use of a stimulus that was difficult to detect led to behavioral errors, which allowed us to compare VTA activity between trials with correct and incorrect stimulus-guided choices. We found a sustained increase in firing rate of both putative dopamine and GABA neurons during the pre-CS period of correct and incorrect trials. The poststimulus phasic response, however, was absent on incorrect trials, suggesting that the stimulus-evoked phasic response of dopamine neurons may relate to stimulus detection. The prestimulus activation of VTA neurons may modulate cortical systems that represent internal states of stimulus expectation and provide a mechanism for dopamine neurotransmission to influence preparatory attention to an expected stimulus.

Keywords: ADHD; attention; dopamine; prefrontal cortex; reward; schizophrenia.

Fig. 1.

A diagram of the timing of events during the behavioral task. The trial initiation, prestimulus period, and conditioned stimulus (CS) were the same across trials. The only exception was that the CS presentation changed between 3 randomly selected locations on the operant chamber wall to which the rat oriented. After the CS, there were 3 possible behavioral choices: nose poke into the correct stimulus location, nose poke into the incorrect stimulus location, or response omission (not shown on the diagram). The timing of these nose poke events shown is approximate. Both correct and incorrect behavioral choices were followed 250 ms later by the trial outcome, either reward or extinguishment of the house light. Note that event times are shown to illustrate important points and are not drawn to scale. Analysis focused on firing rate changes during the prestimulus period immediately preceding the CS (−4,000 to 0 ms) and during the poststimulus period (0 to 240 ms). The behavioral response and the outcome occur well after (>500 ms) the CS analysis window.

Fig. 2.

Extracellular recordings of single units were made in the ventral tegmental area (VTA). The drawing indicates the location of recordings from each rat in relation to bregma.

Fig. 3.

Waveforms, spike rasters, and perievent time histograms for firing rate (Hz) are presented for 2 single units. These units significantly increased their firing rate during the prestimulus period. Stimulus onset is at t = 0 s.

Fig. 4.

Sustained activation of VTA single units before stimulus onset. A: the largest proportion of units (n = 29 of 123 total) was activated during correct trials, whereas the proportion of significantly activated units was reduced during incorrect and omission trials (**P < 0.01). Significance was determined using Z score (see materials and methods for details). B: mean normalized firing rate across units that were significantly activated before the stimulus is plotted (250-ms bins). The shading indicates SE. Stimulus onset is at t = 0 s. The magnitude of change was largest during correct trials and reduced during the other trial types (P < 0.0001). During omission trials, the firing rate did not change. The inset shows that nonnormalized firing rate was highly variable across single units. The x-axis shows the mean rate from the prestimulus period (−6 to 0 s), and the y-axis shows the number of units. C: mean normalized firing rate of the remaining nonresponsive units (n = 94) is plotted with shading indicating SE from the mean. These units did not respond during the prestimulus period. However, the population responded after stimulus onset. D–F: units were split into groups with different firing rates to reduce variability in the population mean perievent time histograms. Top: mean and SE of firing rate (Hz) in 250-ms bins. Bottom: individual units on the y-axis and time around stimulus onset on the x-axis. Units were activated for a sustained period of multiple seconds before stimulus onset. Note that 1 unit with a mean firing rate of ∼20 Hz was excluded from the plots to reduce variability but is included in the statistical tests (ANOVA) reported in the results. VTA unit activation was observed during both correct and incorrect trials, but a change in firing rate did not occur during omission trials.

Fig. 5.

Prestimulus increases in firing rate were aligned to the subsequent stimulus-guided nose poke. A and B: the latency to make a stimulus-guided nose poke differed between trial types. During correct trials (orange, A), rats responded with a mean latency of 0.441 s after stimulus onset, whereas latency occurred 1.100 s after stimulus onset during incorrect trials (green, B). C: mean normalized firing rate of prestimulus responsive neurons aligned to stimulus-guided nose poke onset (at t = 0 s, vertical black line). The shading indicates SE. The orange and green vertical dotted lines mark the average time of stimulus onset preceding the nose poke for correct trials (orange) and incorrect trials (green). Units increased firing rate before stimulus onset and continued to fire at an elevated rate until nose poke onset. The change in firing rate was greater during correct trials (orange line). D: prestimulus responsive neurons did not respond to nose poking events that were not stimulus-guided (i.e., premature nose pokes during the prestimulus period). These data suggest that VTA neurons respond during stimulus expectancy and continue until the stimulus-guided action, but do not respond to actions alone.

Fig. 6.

VTA single units were classified as putative dopamine and GABA neurons on the basis of firing rate and waveform duration. A: putative dopamine neurons were characterized by long waveform duration, whereas putative GABA neurons were characterized by short waveform duration. B: firing rate and waveform duration were used to classify single units into 2 groups.

Fig. 7.

All three groups of VTA neurons (dopamine, GABA, and other) contained neurons that were activated during the prestimulus period. A: similar proportions of each putative neuron group significantly increased firing rate during the prestimulus period (correct trials). DA, dopamine. B: units from all 3 groups of putative neuronal types were activated. The mean normalized firing rate across all responsive units is plotted, and the shading indicates SE.

Fig. 8.

VTA units exhibited a phasic activation after stimulus onset that was significantly different between different types of stimulus-guided behavioral response. A: the proportion of VTA single units that had a phasic response after stimulus onset was the largest on correct trials and reduced on the other trial types (**P < 0.001). B: units from the group of putative dopamine neurons had a phasic response after stimulus onset, whereas other groups of neurons were not activated. C: the normalized (Z score) magnitude of phasic response of dopamine neurons was related to the subsequent stimulus-guided response. The response was not related to the instrumental response or to the trial outcome, both of which occurred well after stimulus onset (>500 ms). The mean normalized firing rate of units from the putative dopamine neuronal group that is activated after stimulus onset (t = 0 s) is plotted in 20-ms bins. The shading indicates SE. The largest activation was on correct trials, and the magnitude was reduced during other trials (P < 0.0001). D: the phasic response for each single unit (including 1 putative GABA neuron, plotted at y = 1) that is significantly activated after stimulus onset (t = 0 s) is plotted across a large time window that illustrates firing rate during both the pre- and poststimulus time periods. The normalized firing rate of all single units that were significantly activated after stimulus onset is displayed on the y-axis, and time around stimulus onset (in 20-ms bins) is displayed on the x-axis. A Z score >2 (yellow) signified a significant increase in firing rate from each single unit's baseline firing rate. These neurons did not have a seconds-long activation before the stimulus onset; rather, they had a phasic activation within 250 ms after the stimulus onset. E: the firing rate (Hz) of neurons represented in A–D is also shown nonnormalized across a large time window including both pre- and poststimulus periods (300-ms bins). The phasic response was significantly reduced during error trials.