Null results of oxytocin and vasopressin administration on mentalizing in a large fMRI sample: evidence from a randomized controlled trial

Abstract

Background: Although potential links between oxytocin (OT), vasopressin (AVP), and social cognition are well-grounded theoretically, most studies have included all male samples, and few have demonstrated consistent effects of either neuropeptide on mentalizing (i.e. understanding the mental states of others). To understand the potential of either neuropeptide as a pharmacological treatment for individuals with impairments in social cognition, it is important to demonstrate the beneficial effects of OT and AVP on mentalizing in healthy individuals.

Methods: In the present randomized, double-blind, placebo-controlled study (n = 186) of healthy individuals, we examined the effects of OT and AVP administration on behavioral responses and neural activity in response to a mentalizing task.

Results: Relative to placebo, neither drug showed an effect on task reaction time or accuracy, nor on whole-brain neural activation or functional connectivity observed within brain networks associated with mentalizing. Exploratory analyses included several variables previously shown to moderate OT's effects on social processes (e.g., self-reported empathy, alexithymia) but resulted in no significant interaction effects.

Conclusions: Results add to a growing literature demonstrating that intranasal administration of OT and AVP may have a more limited effect on social cognition, at both the behavioral and neural level, than initially assumed. Randomized controlled trial registrations: ClinicalTrials.gov; NCT02393443; NCT02393456; NCT02394054.

Keywords: fMRI; functional connectivity; mentalizing; oxytocin; theory of mind; vasopressin.

Fig. 1.

Description of Why/How Task. Following a fixation cross presented for an average of 9 s, each of the 16 experimental blocks began with a prompt question shown for 2.1 s followed by a blank screen lasting 0.15 s before presenting a sequence of eight trial images. Participants were given a max of 2.2 s to respond to each image, and a reminder prompt lasting 0.3 s was shown between each image. Each pre-block prompt began with ‘Is the person’ followed by a descriptive phrase specific to each question. This same phrase was then shown as a reminder between each trial. Recorded BOLD signal was analyzed in a variable epoch manner beginning from the onset of the first image to the offset of the final image of the block. The Why prompts described either intentions inferred from a person's bodily actions (e.g. ‘helping someone’) or emotional states inferred from facial expressions (e.g. ‘proud of themselves’). The How prompts described the physical mechanics of body actions (e.g. ‘lifting something’) or facial orientations (e.g. ‘opening their mouth’). The same set of images were used for both Why and How trials, and participants responded ‘yes’ or ‘no’ with their index or middle finger to indicate whether the person(s) in each image demonstrated that mental state or were performing the action stated in the prompt. On average, the task lasted approximately 4.9 min per person. See online Supplemental Information for additional prompt examples and task information.

Fig. 2.

Differences between OT, AVP, and placebo on accuracy and reaction time. (a) Accuracy for Why and How trials, (b) reaction time for Why and How trials. No significant behavioral differences in accuracy or reaction time were observed for OT and AVP v. placebo in either the Why or How trials. Error bars represent standard error. p < 0.001***, p < 0.01**, p < 0.05*, p < 0.10♱, n.s. = not significant.

Fig. 3.

Differences in neural activity found via t tests contrasting the effects of either OT or AVP v. placebo. (a) Neural activation of Why v. How, (b) neural activation of drug condition for Why v. How. (a) The contrast of Why v. How activation collapsed across all conditions shows robust activation in the mentalizing network ROI for Why trials and mirror network ROI for How trials. Warmer colors correspond to greater activation during Why trials and cooler colors to greater activation during How trials. (b) The graph on the right shows there is no significant difference in neural activation for either OT or AVP v. placebo (p > 0.05) in the mentalizing network ROI for Why trials and mirror network ROI for How trials. The maps on the left show no significant activation (p < 0.001 uncorrected) for either drug condition. Error bars represent standard error. p < 0.001***, p < 0.01**, p < 0.05*, p < 0.10♱, n.s. = not significant.

Fig. 4.

Twenty ROIs chosen from Spunt and Adolphs (2014). (a) Significant functional connectivity changes in the Why/How contrast for OT v. placebo. (b) Significant functional connectivity changes in the Why/How contrast tasks for AVP v. placebo. Note. The ROIs used are very similar to the significant activations shown in Fig. 3a (Why v. How contrast). (a) No significant difference in functional connectivity during Why compared to How trials for OT v. placebo. (b) No significant difference in functional connectivity during Why compared to How trials for AVP (only females) v. placebo (only females). Green squares for p > 0.05, yellow squares for p < 0.05 uncorrected for multiple comparisons, red squares for p < 0.05 FDR corrected, and gray squares for correlations of ROI with itself. PCC, posterior cingulate cortex/precuneus; mPFC, medial prefrontal cortex, vmPFC, ventromedial prefrontal cortex; LpMFG, left posterior middle frontal gyrus; LTPJ, left temporoparietal junction; RTPJ, right temporoparietal junction; RSTS, right superior temporal sulcus; LSTS, left superior temporal sulcus; LvlPFC, left ventral lateral prefrontal cortex; RvlPFC, right ventral lateral prefrontal cortex; LIPL, left intraparietal lobule; RIPL, right intraparietal lobule; LpPMC, left posterior premotor cortex; RpPMC, right posterior premotor cortex; LpIFG, left posterior inferior frontal gyrus; RpIFG, right posterior inferior frontal gyrus; LpMTG, left posterior middle temporal gyrus; RpMTG, right posterior middle temporal gyrus; LSPL, left superior parietal lobule; RSPL, right superior parietal lobule.