Acute effects of methylphenidate, modafinil and MDMA on negative emotion processing

Methylphenidate (MPH) and modafinil are stimulants prescribed for attention-deficit hyperactivity disorder and narcolepsy, respectively, but are also used as cognitive enhancers by healthy people. The authors note that while their cognitive effects have been investigated, less is known about how these drugs modulate negative emotion processing, an important consideration because alterations in processing of negative facial expressions—particularly fearful faces—have been linked to anxiety and mood disorders and are mediated by regions such as the amygdala. Schmidt and colleagues therefore set out to compare the acute effects of a single dose of MPH (60 mg) and modafinil (600 mg) on neural responses to fearful faces in healthy volunteers, using 125 mg MDMA as a positive control because of its established mood-enhancing and prosocial effects. The primary hypothesis was that MPH and modafinil would increase amygdala responses to fearful faces relative to MDMA and placebo, and that such neural effects would relate to changes in negative mood states (state-anxiety, fearfulness, depressiveness). This within-subject, placebo-controlled, crossover fMRI study aimed to detect potential adverse effects on emotion processing despite the drugs' cognitive-enhancing properties.

The study used a double-blind, placebo-controlled, randomised crossover design with four sessions per participant: 125 mg MDMA, 60 mg methylphenidate, 600 mg modafinil, and placebo. Sessions were counterbalanced with washout periods of at least 7 days. Twenty-four healthy volunteers (12 men, 12 women; mean age 22.6 ± 3.0 years, range 19–29) were recruited; inclusion criteria included age 18–45 and body mass index 18–27 kg/m2. Exclusion criteria included personal or first-degree family history of psychiatric disorder, significant physical illness, heavy tobacco use (>10 cigarettes/day), or extensive lifetime illicit drug use (>5 times), with occasional cannabis allowed. Female participants used oral contraceptives and were tested during the follicular phase by self-report. All participants provided informed consent and the study was approved by relevant ethics and regulatory bodies. Drugs were administered orally at 9:45 am and fMRI scanning occurred between 11:15 am and 12:15 pm to capture peak drug effects; each session lasted about 7 hours. Negative emotional states were assessed with the State-Trait Anxiety Inventory (STAI) and the Adjective Mood Rating Scale at 75 and 150 minutes posttreatment, and these two posttreatment measures were averaged for analysis. Facial emotion recognition was measured with a Facial Emotion Recognition Task (FERT) at 150 minutes postdose; the FERT used morphed faces expressing happiness, sadness, anger, or fear (0–100% in 10% steps), and the main outcome was accuracy for fearful relative to neutral faces. The fMRI task assessed implicit processing of fearful faces in a 6-minute event-related design: participants viewed ten identities showing 50% or 100% fear or neutral expressions, producing 60 trials, and performed a gender decision to maintain attention. Imaging was acquired on a 3-T scanner (TR 2.5 s, TE 28 ms, 3 x 3 x 3 mm resolution) and preprocessed in SPM8. Volumes with gross artefacts were inspected; six corrupted volumes (all after MDMA) were replaced by interpolation and participants with >10% corrupted volumes were excluded (n = 2), yielding a final sample of 22 for imaging analyses. First-level contrasts contrasted 50% and 100% fearful faces against neutral, with the primary focus on the 100% fearful versus neutral contrast. Second-level analyses used a within-subject ANOVA with drug order as a covariate of no interest. Whole-brain inference used cluster-level family-wise error (FWE) correction at P < .05 with an initial voxel-forming threshold of P < .001. The amygdala was a priori defined as a region of interest using coordinates from a prior meta-analysis and analysed with small-volume correction (8-mm spheres) at voxel-level P < .05 FWE. Behavioural measures (negative mood and FERT) were analysed with repeated-measures ANOVA and Bonferroni posthoc tests where appropriate. The relationship between drug-induced neural activation and negative emotion measures was tested by including fearfulness, depressiveness, and state-anxiety as covariates in second-level models.

Two participants were excluded from imaging analyses for excessive corrupted volumes, leaving 22 subjects in the final sample. On negative emotional state measures, there was a trend-level main effect of treatment for state-anxiety (F = 2.592, P = .083), with a trend for higher state-anxiety after MPH than after MDMA (P = .073). No significant treatment effects were found for self-reported fearfulness (F = 1.259, P = .317) or depressiveness (F = 1.129, P = .362). On the FERT, there was a significant main effect of treatment for recognition accuracy of fearful relative to neutral faces (F = 3.679, P = .030). Posthoc testing showed that recognition of fearful faces was significantly better after modafinil than after MDMA (P = .039). The authors report no significant relationship between recognition performance and negative emotional states following modafinil. Averaged across treatments, fearful versus neutral faces activated a distributed network including the amygdala, fusiform gyrus, anterior cingulate and orbitofrontal cortices, calcarine sulcus, dorsal striatum, insula, and inferior frontal gyrus (all cluster-level FWE-corrected). Treatment effects during fearful face processing differed across conditions in the left amygdala (small-volume peak-level FWE-corrected), right amygdala, right putamen, left pallidum, and thalamus (cluster-level FWE-corrected). Specifically, modafinil increased activation relative to placebo in the bilateral amygdala (small-volume FWE-corrected), anterior cingulate cortex, right putamen, pallidum and supplementary motor area, and left pallidum, caudate and thalamus (cluster-level FWE-corrected). Modafinil also produced greater activation than MDMA in the right amygdala (small-volume FWE-corrected). Given the modafinil-related activation increases, the investigators tested brain–behaviour relationships: after modafinil intake, activation in the right middle and inferior frontal gyrus correlated positively with self-reported fearfulness, and bilateral middle and inferior frontal gyrus activation correlated positively with reported depressiveness. No significant relationships were found between state-anxiety and brain activation following modafinil administration.

Schmidt and colleagues interpret their findings as showing that a single relatively high dose of modafinil (600 mg) increases neural activation to fearful faces within a limbic–cortical–striatal–pallidal–thalamic circuit, including the amygdala, compared with placebo. They also highlight that modafinil increased amygdala activation compared with MDMA, and that frontal activation in response to fearful faces under modafinil correlated with subjective fearfulness and depressiveness. The authors place these results against prior literature: some studies have reported increased regional blood flow in arousal- and emotion-related regions after modafinil, whereas other work—using different doses or repeated administration—has reported diminished amygdala responses. They suggest dose and dosing schedule may account for these discrepancies. Although 600 mg modafinil did not significantly increase state-anxiety in their sample, the same dose produced notable somatic adverse effects (insomnia, headache, lack of appetite) in participants and previous studies report dose-related increases in adverse events; the investigators therefore propose that sympathomimetic or somatic anxiety symptoms could drive the observed increases in fear-associated brain activation. Mechanistically, the authors speculate that modafinil’s effects may involve reductions in GABAergic neurotransmission—modafinil is a weak dopamine and noradrenaline transporter inhibitor but also affects GABA, glutamate and orexin systems—and that reduced GABA tone could disinhibit limbic structures such as the amygdala. By contrast, 60 mg methylphenidate did not change neural responses to fearful faces relative to placebo, which the authors take to suggest that modulation of dopamine and noradrenaline alone (as with MPH) may be insufficient to alter amygdala responsivity in this paradigm. Concerning MDMA (125 mg), the study did not find reduced amygdala responses to fearful faces versus placebo, consistent with some prior reports showing MDMA attenuates responses to angry but not fearful expressions; however, MDMA produced lower recognition accuracy for fearful faces compared with modafinil and promoted prosocial subjective effects in earlier analyses of the same sample. The authors acknowledge several limitations: including multiple drug conditions may have reduced power to detect treatment effects on negative mood; mood assessment could be improved by using standardised instruments such as the Positive and Negative Affect Schedule or Profile of Mood States; the design precluded exploration of dose–response relationships because only single, relatively high doses were tested; reliance on self-report to index menstrual cycle phase is imperfect; and drug-induced changes in neurovascular coupling could confound fMRI measures. They conclude that although 600 mg modafinil can enhance cognitive performance in healthy people, its acute administration at this dose increased neural activation in regions implicated in fear processing and some of these effects related to negative mood measures, so potential adverse effects on emotion processing merit consideration.