The role of transcutaneous vagus nerve stimulation in post-error slowing

People tend to slow down after they commit an error, a phenomenon known as
post-error slowing (PES). Specifically, PES reflects longer reaction times
(RTs) in trials following an error than in trials following a correct response
(Rabbitt, 1966). PES has been observed in a wide range of tasks (Danielmeier &
Ullsperger, 2011), including the Simon task (King et al., 2010; Danielmeier et
al., 2011), the flanker task (Debener et al., 2005; Krämer et al., 2007) and
the Stroop task (Gehring and Fencsik, 2001). PES is particularly pronounced
when the interval between the erroneous response and the subsequent stimulus
(RSI) is short (Jentzsch & Dudschig, 2009; Danielmeier & Ullsperger, 2011), and
when errors are infrequent (Notebaert et al., 2009; Steinborn, Flehmig,
Bratzke, & Schröter, 2012). There is increasing evidence that PES is
independent from other post-error adjustments, such as post-error improvement
in accuracy (PIA). Indeed, recent studies have shown that PES and PIA are not
necessarily correlated: with short RSI, PES is associated with decrease in PIA
while with long RSI, PES is associated with PIA improvements (Danielmeier &
Ullsperger, 2011). With long RSI, it has been suggested that PES is related to
cognitive control processes associated with the error (Ridderinkhof et al.,
2004): a sort of compensatory control mechanism aiming at improving subsequent
performance. With short RSI, PES is supposed to reflect, instead, an
attentional re-orienting (i.e., an orienting response following infrequent
(and, thus, motivationally salient) events like errors; Notebaert et al.,
2009).

Ullsperger and colleagues (2010) have suggested that slowing after unexpected
errors or negative feedback is related to the activity of the neuromodulatory
locus coeruleus-norepinephrine (LC-NE) system. Very recently, Colzato and
colleagues (2013) have demonstrated the dependency of an individual*s size of
PES on the DBH5*-ins/del polymorphism*a variation in the DBH gene associated
with the production of the enzyme dopamine beta-hydroxylase, which catalyzes
the conversion of dopamine to NE. Indeed, DBH5*-ins/del heterozygotes
(associated with average level of plasma DβH activity) showed increased PES
compared to del/del homozygotes and ins/ins homozygotes (associated with low
and high level of plasma DβH activity, respectively).

The proposed study examines the effects of vagus nerve stimulation (VNS) on
PES. VNS affects various parts of the brain including neural activity in the
LC-NE system. In rats, it has been demonstrated that VNS leads to an
intensity-dependent increase in brain NE in response to stimulation of the left
vagus nerve (Raedt et al., 2011; Roosevelt et al., 2006). Animal research by
Roosevelt et al. (2006) showed that stimulation at a current strength of 0.0 mA
and 0.25 mA did not affect the NE concentrations in the hippocampus or the
cortex, while 0.5 mA stimulation significantly increased the NE concentrations
in the hippocampus (bilateral). These increases in NE are transient and return
to base-line levels when the stimulation is stopped and the vagus nerve is not
being stimulated (Roosevelt et al., 2006).

Traditionally, VNS has been performed by the implantation of a neurostimulating
device connected to an electrode located along the cervical branch of the vagus
nerve. In order to minimize adverse effects of this procedure such as coughing
during stimulation, croakiness, general operational and anesthesiological
risks, and high costs, a new non-invasive neurostimulating device has been
developed for transcutaneous stimulation of the afferent auricular branch of
the vagus nerve (ABVN) located medial of the tragus at the entry of the
acoustic meatus (tVNS®) (Kreuzer et al., 2012). The tVNS® device received CE
approval in 2010 (CE1275). CE marking is an indication that a medical device
complies with essential health and safety requirements. Transcutaneous vagus
nerve stimulation (tVNS) targets the cutaneous receptive field of the ABVN at
the outer ear (Ellrich, 2011).

In the proposed study we will test detailed predictions about the effect of
tVNS on PES in one cognitive task using short RSI and in a second cognitive
task using long RSI with combined pupillary response. Several studies suggest
that pupil diameter closely track the time course of LC activity. Rajkowski,
Kubiak, and Aston-Jones (1993), for example, found a strong correlation in
monkeys between baseline pupil diameter and tonic LC firing rate over the
course of 90 min of performance in a target-detection task. Furthermore, a
recent study that investigated how pupil diameter is related to experimental
manipulations of task-related utility and behavioral indices of task
(dis)engagement showed that pupil diameter varied in a way consistent with
predicted LC dynamics (Gilzenrat et al., 2010). Inspired by the recent evidence
that pupil diameter might be used as an indirect index of LC activity, using
long RSI, we will measure participants* pupil diameter while they perform an
auditory four-choice reaction task (CRT). In this task, adapted from Yordanova
and colleagues (2004), the auditory stimuli are the letters A, E, I and O,
which will be presented via external speakers. The participant has to respond
to the letters A, E, I, and O with the left middle, left index, right index,
and right middle finger, respectively. We expect active tVNS, compared to sham
stimulation, to increase pupil diameter, PES and PIA.

With short RSI, the participants will perform a modified version of the flanker
task (Eriksen and Eriksen, 1974), in which a target letter (*H*, *K*, *C* or
*S*) will be flanked by three identical flanker letters (*H*, *K*, *C* or *S*)
on each side. Participants have to classify the target letter by giving one of
two left-hand responses or one of two right-hand responses. In this cognitive
task we will not be able to combine pupillary response because there is not
enough time for the pupil to return to baseline before the next trial will be
presented. In contrast to the long RSI task, we expect active tVNS, compared to
sham stimulation, to increase PES but to decrease PIA.

Our intended experiment, taking into account a long and short RSI, allows us to
further understand the functional role of PES and its neuromodular
underpinnings mediated via LC dynamics. In contrast to our previous genetic
correlative study (Colzato et al., 2013), we will be able to investigate
whether the LC-NE system (through its connection to the vagus nerve) plays a
causal role in the formation of PES and related post-error adaptations such as
PIA.

There are large individual differences in the cognitive effects that we intend
to investigate, which complicates an interpretation of the results.
Personality/mood questionnaires are a source of critical information about
individual differences that can help explain the variance in the effect of the
tVNS on our experimental measures.

Primary Objective: to ascertain whether tVNS affect performance (as measured by
modulation of reaction times and/or error rates) in two cognitive tasks
designed to examine PES and PIA, as well as specific components of pupillary
response that will be used as neural correlates of LC-NE system activity.

Secondary Objective(s): to explore possible individual differences in the
behavioural and pupillary response indices of PES by means of personality/mood
questionnaires. Several personality trait questionnaires will be used and also
a state questionnaire (the affect grid) to gain insight in participants* levels
of arousal, but also in their current mood and how it is affected by tVNS.

The proposed study uses a double-blind, pseudo-randomized, sham-controlled
cross-over design. Participants* task performance and pupillary response are
measured in two sessions, once following active stimulation, a constant current
of 0.5 or 1.0mA (between-subject manipulation) delivered for 75 minutes, once
following, for the same current intensity and duration, earlobe sham
stimulation, in a counterbalanced order.



We will test 48 participants, twelve with each possible order of treatments.
While the pupillary response is measured, participants perform two cognitive
computer tasks (lasting about 25 minutes each; cf. Section 7.3); prior to
pupillary response measurements, participants complete a number of
questionnaires (which will take less than 20 minutes in total). One session
will last about one hour and thirty minutes.

tVNS NEUROSTIMULATING DEVICE
A tVNS instrument consisting of two titan electrodes mounted on a gel frame and
connected to a wired neurostimulating device (CM02, Cerbomed, Erlangen,
Germany) will be used. The tVNS® device will be programmed to a stimulus
intensity at 0.5 or 1.0 mA with a stimulation frequency of 25 Hz. Stimulation
will be active for 30 sec, followed by a break of 30 sec. Following Kraus and
colleagues (2007), in the sham condition, the stimulation electrodes will be
attached to the center of the left ear lobe instead of the outer auditory canal
for stimulation. Since efferent fibers of the vagus nerve modulate cardiac
function, cardiac safety has always been a concern in the therapeutic use of
vagus nerve stimulation (Cristancho et al., 2011). Efferent vagal fibers to the
heart are supposed to be located on the right side (Nemeroff et al., 2006). In
order to avoid cardiac side effects, electrode placement is performed on the
left side in treatment of central nervous diseases (Nemeroff et al., 2006). On
the left side, a clinical trial show no arrhythmic effects of tVNS (Kreuzer et
al., 2012).

Computerized tasks and paper-and-pencil tests
There are no risks associated with the performance of cognitive computer or
paper-and-pencil tasks except the occasional possibility of some frustration
with poor performance or fatigue.

Pupillary response
Pupillary response recording is a standard procedure used in hospitals and
universities across the country. This procedure is completely not invasive and
does not involve risk.

tVNS
Previous studies have used comparable or higher duration and intensity of
stimulation without reporting significant adverse side-effects (cf. Summary).
Thus, the risks associated with a single stimulation are considered minimal.
However, to reduce the remaining risks, strict prescreening procedures (section
4.3) and safety procedures (sections 5.1 and 8.2) are implemented.