Neural Mechanisms of the Freeze-to-Fight Transition

Upon encountering a threat, a state of attentive immobility called the freeze
response can be evoked. This response has been argued to serve various
functions: to avoid detection by a predator or triggering an attack, to prepare
for the fight or flight response, and to accumulate information determining
whether to break out freeze, and if so whether the fight or the flight response
will be selected. During freeze parasympathetic activity is dominant, causing
for instance a decrease in heart rate termed *fear bradycardia*. The transition
from passive freeze to active fight-or-flight involves sympathetic activation
and a heart rate increase (Fanselow, 1994). The ability to transition from
passive freeze to active fight/flight when appropriate is critical for survival
(Blanchard ea, 2011). How does an organism decide that it is time for an active
response? Recent observations suggest that the apparent rules of freeze and
fight/flight in animal may also apply to human defensive responses: 1) Specific
defenses associated with specific threats are shared by lab rats and humans; 2)
Specific components of freeze and fight-or-flight are differentially affected
by distinct pharmacological interventions; 3) Outcome of freeze-related risk
assessment (fight-or-flight) is stable across new and old situations in men and
rodents (Blanchard ea 2011; Korte ea, 2005).

The neural mechanisms of transitioning from freeze to defensive action in
humans remain largely unknown. Studies in rodents have shown that the amygdala
plays a key role in orchestrating defensive behavior and transitioning between
defensive modes (Ledoux, 1988). Whereas freeze involves direct projections from
the amygdala to the brainstem (periaquaductal gray: PAG), active (fight/flight)
responses depend on amygdala projection to the ventral striatum, were
dopaminergic projections facilitate active fight/flight responses (Ledoux,
1988). Testosterone has been suggested to promote the transition from freeze to
active fight/flight by acting on those dopaminergic projections in the striatum
(De Souza ea, 2009; Hermans ea 2010). The aPFC is involved in regulating the
amygdala and striatum-driven actions via top-down control (Volman ea, 2011).
The study of the role of these regions in freeze-fight transitions in humans
has been thwarted by technical challenges. The size and location of the
periaquaductal gray (PAG), that plays a central role in models of defensive
reactions, preclude the use of traditional fMRI protocols, as the signal may be
confounded by motion artefacts associated with heart pulse affecting local
blood vessels. However in a recent pilot study, it has been shown that these
artefacts can be controlled for, neural correlates of fear bradycardia during
aversive versus neutral picture viewing can be detected. This showed a negative
correlation between heart rate frequency and PAG activity. In addition, there
was strong amygdala-brainstem connectivity associated with the fear
bradycardia, providing the first evidence that similar amygdala-brainstem
projections may facilitate freeze-like behavior in humans as in animals. These
findings lay the grounds for investigations of the neural correlates of human
freeze behavior, and most importantly, of the transition from freeze to active
defense responses.

Primary Objective: The primary objective of the current study is to measure the
neural activation involved in transitioning from freeze to fight, in healthy
subjects.
Secondary Objective(s): To measure connectivity patterns related to freeze and
fight-preparation; to determine correlations between various physiological
measures (including heart rate, body sway, and brain activation) during
different task conditions; and to help determine the optimal version of a
freeze-to-fight/flight task for a future longitudinal study.

The study design for the primary research objective is a within-subject full
factorial design with healthy adult subjects. Subjects will perform a Shooting
Task during a one-hour behavioural session and a subsequent one-hour fMRI
scanning session. The Shooting Task will involve two primary experimental
factors: a freeze versus no-freeze state, and movement preparation versus no
movement preparation. The study is expected to be completed within six months,
and will be performed at the Donders Centre for Cognitive Neuroimaging.

Subjects will undergo the following measurement procedure. First, subjects will
receive and information brochure and, after signing an informed consent form,
be screened to determine if they can participate in fMRI research. At the lab,
subjects will fill in the questionnaires. They will then practice the Shooting
Task outside the scanner, while standing on a balance board to measure body
sway. After that the scanning procedure starts, consisting of an anatomical
scan and functional scans during performance of the Shooting Task. When the
subject is *shot* during the Shooting Task, out of the scanner a brief burst of
white noise will be presented, and in the scanner a mild electric shock will be
presented.

Because of the secondary aim of piloting versions of the task for an upcoming
longitudinal study, we aim to pilot three versions of the Shooting Task.

The burden consists of the time and effort involved in performing the tasks
during the measurement session, and experiencing the aversive stimuli (brief
bursts of loud noise and electric shocks). There is no expected benefit beyond
the compensation and personal interest (as many subjects will be students).