Figure 4.

Pharmacological manipulation of the serotonergic system prolongs time to act as a function of the average reward rate of the environment

(A–C) The effect of citalopram (treatment) and placebo (control) on observed actTime. Increasing systemic serotonin levels promoted waiting before responding (A). This effect, however, was not observed when using the combined effect of the immediate contextual factors to predict time to act (deterministic actTime; B) but showed a significant interaction with ITI, such that the effect of citalopram on observed actTime was stronger during long ITI compared with short ITI blocks (C).

(D) Interaction coefficients between drug administration (treatment versus control) and contextual factors in the present and recent past trials.

(E) There was no significant difference in the percentage of non-responded trails between treatment and control conditions.

(F) The effect of citalopram on observed actTime was sensitive to the dose of the administered drug. Intv. is the drug administration group (treatment versus control). In (A), (B), and (E), the gray column is the mean across animals, error bars are the SEM across animals, and each line is data from individual animals. In (C) and (F), error bars show SEM across data. In (D), error bars are standard error of the estimated coefficients. ^∗p < 0.05, ^∗∗p < 0.01. Figure S3 and Table S1 illustrate Cox regression coefficients, the dosing schedule, and effect of drug on 5-HT levels in blood. See also Figure S5.