(C) PLOS One
This story was originally published by PLOS One and is unaltered.
 . . . . . . . . . .



Reduced serotonergic transmission alters sensitivity to cost and reward via 5-HT1A and 5-HT1B receptors in monkeys [1]

['Yukiko Hori', 'Department Of Functional Brain Imaging', 'National Institutes For Quantum Science', 'Technology', 'Chiba', 'Koki Mimura', 'Research Center For Medical', 'Health Data Science', 'The Institute Of Statistical Mathematics', 'Tokyo']

Date: 2024-01 

Effects of 5-HT depletion on incentive

We first determined how much 5-HT is depleted by preventing its synthesis with pCPA. We repeatedly injected pCPA (150 mg/kg, s.c.) over the course of 2 days, which resulted in a decrease in 5-HT metabolites (5-hydroxyindoleacetic acid, 5-HIAA) of at least 30% (N = 2, 33% and 64%) in the cerebrospinal fluid (CSF), while the concentration of DA remained unchanged (Fig 1A and 1B). The 30% value was later used when determining the dosages for specific 5-HT receptor-type antagonists.

PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 1. Effect of 5-HT depletion via pCPA on incentive. (A) Schedule of pCPA treatment, behavioral testing, and CSF sampling. (B) Normalized concentration of 5-HT metabolite (5-HIAA) and DA before (baseline) and after pCPA treatment (depletion). (C) Reward-size task. Left: Sequence of events during 1 trial. A monkey initiated a trial by touching the bar in the chair. After 100 ms, a visual cue signaling the amount of reward (1, 2, 4, or 8 drops) that would be delivered was presented at the center of the monitor. After 500 ms, a red target also appeared at the center of the monitor. After a variable interval of 500–1,500 ms, the target turned green, indicating that the monkey could release the bar to receive the reward. If the monkey responded between 200 and 1,000 ms, the target turned blue indicating the trial had been completed correctly. On correct trials, water rewards were delivered immediately. An ITI of 1 s was enforced before the next trial could begin. If the monkey made an error by releasing the bar before the green target appeared, within 200 ms after it appeared, or failed to respond within 1 s, all visual stimuli disappeared, the trial was terminated immediately, and the trial was repeated after the 1-s ITI. After each correct trial, a new cue reward-size pair was picked from the set of 4 at random. Right: Relationship between visual cues and reward size. (D) Representative error rates as a function of reward size for monkey BO. Dotted curves are the best fit of inverse functions (Model #4, S3 Table). (E) Normalized error rate (percent of maximum error rate in 1 drop trial in the control session; mean ± SEM) as a function of reward size for n = 11 sessions collected from 4 monkeys (S2 Table). (F) Schematic illustration of increase in error rate as incentive a is reduced. (G) Box plot of normalized incentive (a) for each treatment condition (n = 11 for each). Each value was normalized to the value of the control condition. (H) Schematic explaining the increase in error rate by e, independent of reward size. (I) Box plot of parameter e (normalized as the ratio of maximum error rate in 1 drop trial in the control session; mean ± SEM) for each treatment condition (n = 11 for each). The data underlying this figure can be found in https://doi.org/10.5281/zenodo.10141750. 5-HIAA, 5-hydroxyindoleacetic acid; CSF, cerebrospinal fluid; DA, dopamine; ITI, inter-trial interval; pCPA, para-chlorophenylalanine. https://doi.org/10.1371/journal.pbio.3002445.g001

We next examined the effects of 5-HT depletion on incentive in 4 monkeys that were not used in the CSF study (S1 and S2 Tables). For this purpose, we used a reward-size task in which the amount of reward was manipulated across trials, but the task requirements (i.e., the costs) remained the same (Fig 1C). For each trial, the monkeys could receive a reward if they released a bar when a visual target changed from red to green. A visual cue at the beginning of each trial indicated the amount of reward they could get (1, 2, 4, or 8 drops). All monkeys had been trained to perform basic color discrimination trials on a cued multi-trial reward-schedule task [17] for more than 3 months. As in previous experiments using a single option task, the required action was very easy, and monkeys could not fail if they actually tried to release the bar at the proper time (the error rate is indeed much lower in the absence of information about costs and benefits) [2]. As in previous experiments in which costs and benefits were manipulated, errors (either releasing the bar too early or too late) were usually observed in small reward trials and/or close to the end of daily sessions [2,18,19]. Note that error trials were repeated with the same cue-reward condition, which prevented the monkeys from skipping unwanted trials. Therefore, bar “errors” were considered to have occurred when the monkeys were not sufficiently motivated to release the bar at a time that would lead to reward. The frequency of error trials is thus a reliable metric for quantifying the influence of motivation on behavior [15,18,19]. Furthermore, we have previously shown that the error rate (E) is inversely related to the reward size (R), which has been formulated with a single free parameter a [2] (Fig 1F), (1)

This inverse relationship was consistently observed in the control condition in all monkeys (e.g., CON in Fig 1D and 1E). After pCPA treatment, error rates increased independently from reward-size-related errors. For example, in monkey BO, the error rate became progressively higher in sessions that followed the first and second treatments, while differences that depended on reward size appeared to remain the same (Fig 1D, pCPA-day 1 and 2). A reward-independent increase in error rate was consistently found in all monkeys tested, as shown in the average plot of the normalized error rate (Fig 1E). We next quantified how much of the increase in errors was related to reduced incentive (i.e., devalued reward) and how much was reward-size independent. These factors can be captured by a decrease in parameter a (reward impact or incentive) of the inverse function and implementation of the intercept e, respectively (Fig 1F and 1H). To quantify the increase in error rate, we compared 5 models that considered these 2 factors as random effects: Model #1, random effect on a; Model #2, random effect on a fixed e; Model #3, independent random effects on both a and e; Model #4, a single normal distribution of random effect on a and e; and Model #5, random effect on e (see S3 Table). Model #4 was selected as the best model with the lowest Bayesian information criterion (BIC) value (S3 Table), indicating that the increase in error rate was explained by simultaneous changes in parameters a and e. Our model-based analysis revealed that, compared with the pre-treatment baseline, a was significantly lower on day 2 of pCPA treatment (one-way ANOVA, main effect of treatment, F (2, 20) = 4.6, p = 0.023; post hoc Tukey HSD, p = 0.025 for pCPA-day2 versus CON; Fig 1G). At the same time, parameter e was significantly higher on pCPA-day 2 (main effect of treatment, F (2, 20) = 4.1, p = 0.031; post hoc Tukey HSD, p = 0.025 for pCPA-day 2 versus CON; Fig 1I). These results suggest that the 5-HT depletion-induced increase in errors can be explained by 2 components: one is reduced incentive (a), and the other is a factor that appears orthogonal to the incentive value (increase in parameter e). We hypothesized that the value-independent component reflected expected cost and subsequent tests supported this interpretation (see Effects of 5-HTR blockade on cost-based motivation below).

Given that reward value (and thus incentive) decreases as animals become satiated [19], we further investigated how the error rate increased along with satiation. In Fig 2A, the average error rates from the normalized data (n = 11) are replotted as a function of normalized cumulative reward (see Materials and methods). As previously shown, overall error rates in the control condition increased for each reward size as the normalized cumulative reward increased. This satiation-dependent change in error rate was commonly observed among the 3 conditions, with the effect being stronger in the 2 posttreatment sessions (pCPA-day1 and -day2) and corresponding to reduced incentive due to 5-HT depletion (cf., reduced a, in Fig 1G). However, we also observed pronounced increases in posttreatment error rates that were independent of satiation level; namely, error rates became higher even during the early phase of the session, presumably when thirst drives were still high (Fig 2A). Indeed, fitting the data to an error-rate model that incorporates the satiation effect (Eq 4) showed that regardless of reward size or satiation level, the error rates were higher in posttreatment sessions (e = 3.3 and 6.6 for pCPA-day1 and -day2, respectively) than in the control session (e = 0.74). This again suggests that 5-HT depletion blunts motivation in a manner partially independent from reduced incentive.

PPT PowerPoint slide

PNG larger image

TIFF original image Download: Fig 2. Effect of 5-HT depletion on impact of satiation on error, RT, error pattern, and RT-error relationship. (A) Error rate (mean ± SEM; n = 11 sessions) as a function of normalized cumulative reward for control and pCPA treatment conditions. Colors indicate reward size. Curves indicate the best-fit model of Eq 4. (B) Mean RT (mean ± SEM) as a function of reward size for control (CON) and pCPA treatment conditions in monkey BO. Colors indicate treatment condition. (C) Rate of early and late errors (mean ± SEM) for control and pCPA-day2 conditions. (D) Early release rate (mean ± SEM) for control and pCPA treatment conditions. (E) Relationship between error rate and mean RT for each reward size in the first and second half of each session under pCPA treatment in monkeys BO and TO, respectively. Colors indicate treatment condition. Colored lines represent the best-fitting linear regression models that explained the data (S4 Table). The data underlying this figure can be found in https://doi.org/10.5281/zenodo.10141750. pCPA, para-chlorophenylalanine; RT, reaction time. https://doi.org/10.1371/journal.pbio.3002445.g002

[END]
---
[1] Url: https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3002445

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/