People with and without anxiety disorders learn to fear a threat equally quickly, according to new research. However, people with anxiety disorders tend to have a harder time learning to stop being afraid on a physiological level (when the threat is gone) compared to healthy individuals. These trends can be detected by monitoring fear-potentiated startle responses.
The study was published in the International Journal of Psychophysiology.
Learning to recognize threats is a capacity that is of extreme importance for survival. However, problems in these processes, such as situations when an individual easily learns to associate certain stimuli with danger but fails to extinguish this association once the stimuli in question are not longer associated with danger, have long been linked to the development and maintenance of pathological anxiety.
Some theorists have linked the fact that anxiety symptoms start developing in late childhood and early adolescence with variations in the maturation of neural circuitry supporting threat learning. This is often explored experimentally by using the so-called differential threat conditioning study plans.
These research studies pair one neutral stimulus with another that is clearly unpleasant or painful (so-called unconditional stimulus) and present another neutral stimulus just by itself. They then compare the fear reaction to the neutral stimulus that was paired with the unpleasant/painful one to the fear reaction (or its absence) to the neutral stimulus that was presented alone.
Scientists most often use two different indicators of fear – skin conductance response and fear-potentiated startle response.
Skin conductance response assesses fear through measuring the conductivity of the skin. When a person is afraid (or more precisely – aroused, either positively or negatively), the sympathetic part of the autonomic nervous system activates sweating, skin becomes more moist, thus increasing conductance (water is a conductor).
The fear startle response is a reflexive physiological reaction to a stimulus one perceives as a threat. It is most often measured by monitoring eye blinks – a person will tend to blink when something he/she perceives as dangerous approaches.
Study author Philip Newsome and his colleagues wanted to explore whether anxiety effects on differential threat conditioning and extinction (learning to stop reacting with fear to a neutral stimulus that one previously associated with danger) can be detected by monitoring a specific type of fear-potentiated startle response.
They reanalyzed a dataset from a 2020 study containing responses of 306 persons. Of the sample, 133 were diagnosed with anxiety or a similar disorder, while the remaining patients were diagnosis-free. The threat conditioning was done by presenting pictures of two women (two neutral stimuli) to participants.
After the first picture was presented, another picture of the same woman was presented soon after, but the woman in this new picture showed fear. This was accompanied by a 95-decibel female scream delivered via headphones. The picture of the other woman was presented just by itself. Each picture was presented 4 times for 8 seconds.
Fear response was measured through fear-potentiated startle. 4-10 psi of compressed air were blown on the forehead of the study participant 5-6 seconds after showing a picture of a women and an electromyogram recorded whether the participant blinked.
The researchers additionally recorded skin conductance responses using two electrodes attached to middle and ring fingers of the participant’s non-dominant hand. Participants also gave ratings of their subjective feeling of fear. They collected data on the treatment outcome of participants diagnosed with anxiety (or a similar disorder). Finally, researchers analyzed data from magnetic resonance images of participants’ brains, but these were available only for 205 participants.
Results showed that both the group of patients with an anxiety diagnosis and the group without a diagnosis successfully learned fear conditioning responses. In other words, they quickly learned to react with fear to the picture that was followed by the loud scream.
This finding was obtained both by monitoring the fear-potentiated startle response and skin conductance. Participants diagnosed with anxiety on average reported being more afraid than diagnosis-free participants. Older participants reported being less afraid.
When learning to extinguish the threat reaction was considered, results of skin conductance response monitoring pointed to similar rates of diminishing response to the picture that used to be accompanied by a loud scream. As trials during which the picture was not followed by the scream proceeded, fear reaction measured through skin conductance gradually decreased.
However, when fear-potentiated startle responses were examined, results showed that participants with the an anxiety disorder diagnosis extinguished the fear response much slower than participants from the diagnosis-free group. Even at the end of the procedure, the fear-potentiated startle response was higher for the picture that used to be associated with the loud scream. A similar finding was obtained when self-reported fear was compared between the two groups.
“Together, fear-potentiated startle responses, but not skin conductance responses, reveal retainment of threat contingencies in pathological anxiety. These findings suggest that fear-potentiated startle responses and skin conductance responses might capture distinct processes related to conditioned threat responding,” the study authors concluded.
The study gives an important contribution to scientific knowledge on the physiology of fear responses. However, it should be noted that the lack of effects of anxiety on the skin conductance response might have to do with the fear extinction phase being too short. Additionally, it is possible that methods of measuring fear were themselves somewhat unpleasant. This might have affected the validity of fear monitoring data.
The study, “Fear-potentiated startle reveals diminished threat extinction in pathological anxiety”, was authored by Philip Newsome, Sonia G. Ruiz, Andrea L. Gold, Daniel S. Pine, and Rany Abend.