VNS, Acetylcholine and Cortical Plasticity

Acetylcholine (ACh) is a neurotransmitter which was first identified by Henry Hallet Dale in 1914 who observed its interaction with heart tissue. This observation was confirmed by Otto Loewi who decided to call it 'Vagusstoff'. Why was it called this? Because it was found to be released by the vagus nerve.

They both won the 1936 nobel prize in Physiology/Medicine. Why is this important? Recently the vagus nerve has become the focus of a lot of neuroscience attention due to the effect of vagus nerve stimulation. Engineer et al (2011) and Engineer et al (2012) both demonstrated how vagus nerve stimulation can be used to condition the brain and by matching stimulation with stimuli, could effectively treat tinnitus. The research showed a remapping effect of auditory neurons. This was tested only in a mouse model but the results were so promising that a project replicating this is currently being recruited for, except to study this effect in humans.

Porter (2011) has demonstrated how matching vagus nerve stimulation with simple motor tasks allows for a remapping of the motor cortices of the mouse brain. The mechanism of action behind vagus nerve stimulation seems to be unknown.

Stimulation of the vagus nerve may release ACh into the cortex, as observed by Otto Loewi, which somehow causes a cortical plastic effect. Again though, we are faced with the question of How?

Well, ACh has long been linked to synaptic plasticity and acts either by:

Directly enhancing currents through NMDA receptors which have been associated with synaptic plasticity with specifics to memory and learning.

or

Indirectly suppressing adaptation; Neural adaptation, also known as 'up regulation' and 'down regulation' is the process whereby neurons stop firing as a result of constant stimulation allowing regulation to occur.

Along with these observed effects, ACh has been noted to effect heart rate which is something also noted in patients receiving vagus nerve stimulation.

It would definitely seem that rather looking at VNS and cortical plasticity...it's time to look at VNS, causing changes in ACh activity causing plasticity.

Guidance Hypothesis

There has been a lot of research looking at the effect that extrinsic feedback and excessive extrinsic feedback has on learning. Salmoni et al (1984) suggested the guidance hypothesis. This hypothesis suggests that even though recurrent feedback provided during practice is beneficial to the learner in order to choose the correct responses, it blocks the processing of other sources of important information that are essential in order to obtain an internal depiction of the movement task that is capable of generating the movement when the feedback is stopped. Faded schedule feedback was a type of feedback tested to see if this would reduce participant dependency on extrinsic feedback. Winstein et al (1990) performed a study which contained two groups. One group received extrinsic feedback for every trial, while the other group used a faded feedback schedule and had a reduced frequency of feedback (50% of the trials). The results showed that the faded feedback group were able to sustain performance during practice at the same level as the group who had feedback on every trial. However when no feedback tests were administered on both groups after 5 minutes and 24 hours, the faded feedback group performed significantly better. By using the faded feedback schedule, participants were able to reduce their dependency on the feedback which enhanced their capability to create the necessary movement pattern when the feedback was taken away. Other researchers have also performed experiments which have shown the beneficial effects that faded feedback can have (e.g. Lee et al 1990, Weeks et al 1993).

The guidance hypothesis therefore suggests that the faded feedback schedule should cause the participant to process other sources of intrinsic information that assists the progression of an internal depiction which is capable of supporting performance when feedback is taken away. If this hypothesis is true, participants should be able to decrease the time that is needed to develop an internal depiction of the task, from days to just minutes. This suggests a faster way of learning.

Kovacs, A. J., Buchanan, J. J., & Shea, C. H. (2009a). Bimanual 1:1 with 90° continuous phase

Kovac et al (2009a) performed a study looking at the effects of Lissajous feedback. 20 participants were asked to perform a 1:1 bimanual coordination task with a 90° relative phase. While doing this the participants received constant Lissajous feedback. This feedback was in the custom of a cursor signifying the joint position of the two limbs put over on a Lissajous template portraying the necessary phase relation between the limbs. When the left limb moved it moved the cursor vertically and when the right limb moved it moved the cursor horizontally. The results showed that the participants were reasonably effective in executing the coordination pattern that was required (variability and relative phase errors low), after only having 5 minutes of practice. When compared with results from participants who had many days of practice, the level of relative phase errors and variability was in fact lower. In this experiment an auditory metronome was not included and vision of the limbs was not allowed. When a metronome was introduced performance decreased as variability and relative phase errors increased considerably.

Other work by Kovac in 2009 (Kovac et al 2009b) showed that when participants are provided again with constant Lissajous feedback, they can successfully perform relative phases between 30° and 150° using only four minutes of practice at both separate relative phases. This and the study previously discussed, seems to suggest that Lissajous feedback inhibits a faster way of learning.

However a problem with the Kovac studies is that it was proved that participants became too heavily dependent on the Lissajous feedback. When the Lissajous feedback was removed performance decreased as there was higher relative phase error and variability. This outcome proposes that the Lissajous plot with cursor and template provided participants a platform by which they were able to notice their coordination mistakes and perform the required adjustments. This decrease in performance when the Lissajous feedback had been removed suggested that participants had not actually learnt the relative phase but instead had learned to use the constant information provided to perform the necessary coordination patterns. Participants had not developed an internal depiction of the task and therefore were reliant on the constant feedback and when that was removed they struggled with the task. This therefore suggests that constant feedback could indeed hinder learning.

tVNS, Caffeine & Motor Learning - Pilot Results

Vagus nerve stimulation has been shown to increase global neural plasticity in the cortex (Engineer et al, 2011). This type of neural plasticity can be used to facilitate changes in the cortex to treat neurological and psychological disorders (Sclaepfer et al, 2007; George et al, 2008; Uthman et al, 2004 & Bodenlos et al, 2007), and has also been shown to increase consolidation effects when used in learning tasks. These experiments used VNS over a period of weeks and months.

These studies all used iVNS which is a form of VNS which relies upon an implanted electrode. Kraus et al (2004) has showed successfully that tVNS, a method of VNS which utilises nerve fibre connections in the auricular canal and is a reliable method of stimulating the vagus nerve.

Caffeine is a heavily used stimulant which acts on the central nervous system by blocking adenosine receptors. This kind of stimulation has been investigated for the possibility of it having beneficial cognitive effects, specifically on memory and learning. The current research landscape of caffeine and it's effect on cognition is quite noisy. Angelucci et al (1999) has shown that caffeine differentially affects the different stages of memory processing and that its effect depends on the particularities of the task itself. Mednick et al (2008) has shown that motor skill learning is significantly decremented by the use of caffeine compared to a control group, however showed that perceptual learning was significantly increased compared to a control group. Tieges et al (2006) has shown that anticipatory processes are benefited by a caffeine supplement, which aids task switching.

In amongst all of this research lies the question still, can caffeine aid learning when it comes to a task that relies on perception and action. More specifically, the task of producing a 90 degree movement.

Participants were recruited through the use of recruitment posters. 2 male participants were recruited to the tVNS group and female 1 participant was recruited to the caffeine group. Control data was acquired from a previous study with the same design.

Participants took part in an experiment which assessed baseline ability at creating a 90 degree movement on day 1, training with either the stimulation or supplement took place on day 2, a post training assessment took place on day 3 and a retention assessment took place on day 10.

Initial results (see figure above) show that control data outperforms both the vagus nerve stimulation and caffeine group. The control group ability on creating a 90 degree movement increased by 0.32 between baseline and post training whilst the caffeine group increased by only 0.12 and the tVNS group increased by only 0.06. The results of this study show that a short training session on one day may not be enough to create a global plastic effect needed to facilitate learning and may actually inhibit learning to an extent.

After a discussion with both Andrew Wilson and Jim Deuchars on the experiment design and pilot data a decision has been made to move forward with a longer training schedule expanded over various days. Retention will also be assessed 7 days after post training assessment and again 14 days after post training assessment to investigate the consolidating effects of tVNS.