Freezing behavior, a commonly observed defensive stress response, shows relatively high stability over time in animals. Given the relevance of freezing for stress-coping and human psychopathology, it is relevant to know whether freezing behavior is also stable in humans, particularly during adolescence, when most affective symptoms develop. In a prospective longitudinal study, we investigated freezing-like behavior in response to social threat in 75 adolescents at age 14, repeated 3 years later at age 17. We used a well-established method combining electrocardiography (ECG; heart rate) and posturography (body sway) in response to emotional picture-viewing of angry, happy, and neutral faces. We hypothesized that individual differences in freezing-like behavior in response to social threat-operationalized by contrasting angry vs. neutral faces-would be relatively stable over time. Our results indeed showed relative stability between ages 14 and 17 in individual differences in freezing-like behavior in heart rate (r = 0.82), as well as in combined heart rate and body sway measures (r = 0.65). These effects were not specific for the angry vs. neutral contrast; they were also visible in other emotion contrasts. Exploratory analysis in males and females separately showed stability in body sway specifically for angry vs. neutral faces only in females. Together, these results suggest moderate to strong stability in human freezing-like behavior in response to social threat from mid to late adolescence (with exception for the body sway measure in males). This relative stability was not specific for threat-induction and may reflect a general stability that is particularly strong for heart rate. The fact that this relative stability was found over a relatively long time range of 3 years is promising for studies aiming to use freezing-like behavior as a marker for internalizing symptoms in adolescent development.
Keywords: adolescence; freezing-like behavior; prospective longitudinal study; social threat; stability.