Oxford study identifies brain mechanisms behind our sense of control

A study led by the University of Oxford has identified the brain mechanisms that help us work out whether the consequences we experience are caused by our own actions or by circumstances beyond our control.

The study, published today (21 August) in Neuron, found that both confidence in our own decisions and estimates of external circumstances provide important clues. When people were confident in a decision but received an unexpected negative outcome, they were more likely to attribute it to external circumstances. On the other hand, when they were less certain about their decision, they were more likely to consider their own performance as the cause.

Participants made these conclusions partly because they tracked how certain they were about their own performance throughout the assessment even before they obtained any feedback.

Using ultra-high-field brain imaging and targeted non-invasive brain stimulation, the researchers identified two prefrontal-subcortical brain circuits involved in this process. The findings provide new insight into how our sense of control is formed and how it subsequently shapes learning from success and failure.

One of the brain areas the researchers highlight is linked to brain chemicals that are the focus of many treatments for depression. This brain area signals when there are changes in how controllable the world seems to us.

Knowing whether an outcome was caused by us or by something outside our control is fundamental to learning. If an archer misses a target, for example, they need to decide whether to correct their technique or compensate for an external factor such as the wind. Making the wrong attribution could lead them to learn the wrong lesson.”

Dr. Yanhe Liu, Lead Author, Department of Experimental Psychology, University of Oxford

In the study, adult participants completed a decision-making task in which they judged whether there were more blue or orange dots on a screen and rated how confident they were in each choice. They then received feedback on whether their answer was correct.

Crucially, the reliability of this feedback changed. In ‘controllable’ blocks, feedback reflected participants’ actual performance on 90% of trials. In ‘uncontrollable’ blocks, this was true on only 10% of trials, with the remaining feedback generated randomly. Participants were not told which type of block they were in, but had to learn this from experience.

The researchers found that people used their confidence to help work out what had caused an outcome. When participants were highly confident but received unexpected negative feedback, they were more likely to conclude that it had been generated randomly. When they were less confident, they were more likely to attribute negative feedback to their own performance. At the same time, they used these experiences to continually update their estimate of how much control they generally had over the external circumstance.

Using an ultra-high-resolution functional MRI scanner in 22 participants, the researchers traced this process to a circuit connecting the dorsomedial prefrontal cortex (dmPFC) with the dorsal raphe nucleus, a small structure deep in the brain. Activity in the dmPFC tracked participants’ confidence, their estimate of how much control they had, and whether they attributed an outcome to their own performance or to randomness.

A second circuit, connecting another region of the prefrontal cortex with dopamine-associated regions of the midbrain, was linked to how these judgements influenced the brain’s response to feedback.

The researchers then tested the role of the dmPFC directly in a separate experiment with 20 participants. Temporarily disrupting activity in this region using non-invasive magnetic brain stimulation made participants slower to work out whether the feedback they were receiving genuinely reflected their performance or was mostly random. This provides strong evidence that the dmPFC is important for judging what is and is not within our control.

Co-author Professor Matthew Rushworth (Department of Experimental Biology, University of Oxford) said: “Our sense of control has profound consequences for how we behave. If we believe an outcome was caused by our own actions, it makes sense to change those actions next time. If we believe it was caused by something outside our control, the appropriate response may be very different. This study begins to reveal the brain circuitry that enables us to make that distinction.”

The findings could ultimately help researchers better understand why people’s perceptions of control can sometimes become distorted. Altered perceptions of control and causal attribution have been associated with conditions including depression, anxiety and schizophrenia, although the researchers stress that the present study investigated healthy adults and was not designed to test clinical interventions. Nevertheless, the dorsal raphe nucleus – one of the areas the researchers highlight – is strongly linked to serotonin, which is the focus of some of the most common treatments for depression such as SSRIs.

Dr Liu added: “The ability to distinguish what we can control from what we cannot is central to adapting successfully to the world around us. Understanding the brain mechanisms behind that judgement gives us a foundation for investigating what happens when this process becomes biased or disrupted.”