Imagine a person is terrified of dogs because they once suffered a terrible bite. Following long-established techniques, their psychologist might gradually expose them to dogs in a safe setting, until their fear gradually faded away. This “exposure therapy” can be effective but it has some serious drawbacks, including the fact that the person might at first find it traumatic to be close to dogs again.
What if there were a way to remove this person’s fear of dogs at a subconscious level, without the need for any traumatic exposure? Such an approach has now come much closer to clinical reality thanks to a new study reported recently in Nature Human Behaviour. The findings suggest that neurofeedback can be used to unlearn a fear by pairing relevant non-conscious neural activity with a reward, such as money. Significant technical hurdles remain before this becomes a real-life treatment, but it’s an exciting breakthrough.
Ai Koizumi at ATR Computational Neuroscience Laboratories in Japan and her colleagues trained 17 participants (11 men) to fear two colourful grating patterns presented on a computer screen by repeatedly pairing the patterns with safe but unpleasant electric shocks. Toward the end of this session, measures of the participants’ electrodermal response (skin sweatiness) to the sight of the two patterns confirmed that they now found them aversive. Prior to the fear-induction, the researchers had also scanned the participants’ brains while they looked at these two patterns, so the researchers knew what neural activity tended to accompany the mental representation of each one. After the fear-induction process, one of the patterns (the “target”) was chosen as the learned fear to be removed, the other acted as a control.
Next, over three days, the participants undertook several sessions of neural feedback training, during which the researchers monitored their brain activity. This involved the participants looking at a grey grating pattern and then trying to use any mental strategy they liked to try to get a small disc on the screen to increase in size. If the participants succeeded, they were rewarded with small amounts of money.
At the first session, the participants found this challenge impossible, but gradually and subconsciously they improved at the task. Unbeknown to the participants, their success was dependent on their displaying a particular form of brain activity (hence why this was a kind of neurofeedback training). More specifically, what made the disc grow was the participants’ visual cortex exhibiting neural activity previously associated with representing the target grating pattern. In other words, whenever the participants’ brains showed non-conscious neural activity related to the target feared pattern – which they did occasionally – the disc grew, and they were rewarded with money.
During the training, the participants showed no experience of fear, either subjectively or based on their electrodermal response (the sweatiness of their skin), suggesting the fear-related brain activity was entirely non-conscious. And quizzed after the training, it was clear the participants had no idea what had been going on at a subconscious level – they didn’t know what they had been doing mentally to increase the disc size and receive money, and they didn’t know which grating they had been trained not to fear and which one was the control. So while the researchers were busy training the participants’ brains to like one of the patterns that they’d previously learned to fear, the participants themselves were blissfully ignorant of what was going on.
Of course the key test was what happened when, after the neurofeedback training, the researchers subsequently re-exposed the participants to the two colourful grating patterns. As hoped, the participants showed a much weaker fear response (again, based on their skin sweatiness) to the sight of the target grating, as compared with the control grating. In fact, the magnitude of this fear reduction was similar to that seen for conventional exposure therapy. The researchers also scanned the participants’ brains during this testing phase (as they had done during the fear-induction phase) and this showed that the participants’ amygdala activity was now lower in response to the target grating compared with the control grating. The amygdala is involved in fear processing so this provides more evidence that the fear-removal process had worked.
So what are some of the technical hurdles before this kind of approach reaches your local psych clinic? Let’s return to our imaginary individual who is terrified of dogs. For the subconscious neurofeedback training to work, we’d need to know what kind of brain activity pattern this person shows when she encounters dogs, in order that we could pair this activity, at a subconscious level, with a reward. But this dog-related brain activity will be more far more complex than the neural activity associated with a simple grating pattern on a computer screen. And to find out what it looks like, we’d of course need to scan the person’s brain while we exposed him or her to dogs – defeating the whole object of the cunning neurofeedback approach!
There may be a way around this, the researchers suggested – for example, by extrapolating from other people’s dog-related brain activity or by decoding dog-related brain activity based on showing our client subliminal dog images (I’m running with the dog example here but the same principles would apply whatever it is that a client feared). “Despite these challenges,” Koizumi and her colleagues concluded, “the present results hopefully represent an initial step towards a potential new avenue for treatment.”