Depression: semaglutide changes some choices in an experimental task

A task to distinguish different ways of choosing

Participants were assigned to oral semaglutide, with 35 people, or placebo, with 37 people. They completed a task involving choices between four options at the start of the trial and again after 16 weeks of treatment.

The researchers used a computational reinforcement learning model to analyze successive choices. The aim was to distinguish several behaviors: favoring the option with the highest expected reward, exploring a more uncertain option, or repeating the previous choice regardless of its value.

This distinction matters: choosing a known option and exploring an uncertain one are not necessarily opposing behaviors across an entire task. The model seeks to isolate different components of decision-making.

Three changes, but no evidence of better choices

Compared with placebo, semaglutide increased the tendency to choose the option offering the highest expected reward, as well as directed exploration of more uncertain options. It also increased perseveration, meaning the tendency to repeat the previous action regardless of its value.

These findings therefore do not support simply describing the effect as “improved” decision-making. Repeating a choice more often does not, on its own, indicate that the choice is more appropriate. The abstract reports no benefit for everyday decisions.

According to the authors, participants’ baseline characteristics and weight loss modulated the observed effects without fully explaining them. However, this is not enough to identify the mechanism responsible. The involvement of metabolic mechanisms remains a possible interpretation, not a detailed causal demonstration.

A neuropsychiatric avenue with limits still to overcome

Random assignment and the placebo comparator strengthen the interpretation of the observed differences. But this is a secondary analysis of a small sample, with findings derived from a task and modeling. Their clinical significance remains to be established.

The dossier contains only the PubMed abstract. It does not allow examination of the full analyses, adverse effects, funding, or conflicts of interest. These findings justify neither using semaglutide to improve your decisions nor concluding that it has an antidepressant effect.

What we can take away

In this trial, semaglutide changed several experimental components of decision-making in people with depression. Changing a measured behavior does not mean demonstrating a clinical benefit.