Abstract
According to Gray and McNaughton (2000), anxiety is the result of conflict between simultaneously activated approach and avoidance tendencies. The function of the Behavioural Inhibition System (BIS) is to resolve this conflict between goals that are equally activated with regard to strength and timing. Conflict resolution is thought to involve recursive computation in cortico-hippocampal loops resonating at the rodent hippocampal frequency (4-12Hz, “theta”). The BIS theory is based on findings from the animal literature, but recently, attempts have been made to apply it to humans.
Neo, Thurlow, and McNaughton (2011) used the stop signal task (SST) to elicit and measure activation of the BIS in humans. They used a novel analysis, dividing trials so that a high approach-avoidance conflict condition could be compared against the average of non-conflict conditions of approach and avoidance. They found increases in “theta” activation over the right frontal cortex during the high conflict condition. They interpreted this as being consistent with activation of the BIS. The size of this activation was also positively correlated with trait anxiety and with neuroticism across participants.
The overarching goal of the current thesis is to challenge this “theta” activation with two distinct classes of anxiolytic drugs. The drugs are useful for this purpose for two reasons. Firstly, if their effects overlap this can be attributed to their common main anxiolytic effect, if they differ this will be due to their side-effects, which do not overlap (Gray & McNaughton, 2000). Secondly, all classes of anxiolytic drugs reduce hippocampal theta rhythm. If the drugs reduce the “theta” produced in the SST, then parallels can be drawn between humans and animals. The first two experiments of this thesis were preparatory for this final drug experiment. In the first experiment we improved the SST (as used by Neo et al, 2011), such that the conflict specific processing could be more accurately measured. In the second experiment we assessed whether conflict specific processing remained robust over days of testing to see if a within-subject ABCCBA design was feasible for the final drug experiment. We administered the experimental procedure six times to the same participants, and found a conflict effect at 9-10Hz over right frontal cortex only on day one. Thus, in the final experiment we employed a double-blind, randomised, placebo controlled design to challenge this conflict specific processing. Participants were randomly assigned to a triazolam (GABA-A), buspirone (5-HT 1-A), or placebo group. We observed a distinct conflict effect at 9-10Hz over right frontal cortex (at F4 and F8) in the placebo group. The effect was abolished in both of the drug groups.
In summary, we have identified a particular brain signature that is specifically sensitive to anxiolytic drug action. This particular signature has previously been related to high trait anxiety and neuroticism. These results have implications for the development of new classes of anxiolytic drugs, for anchoring BIS personality factors to a direct measure of BIS activation, and for our understanding of anxiolytic processes in humans.