Within task exploratory analysis
In the Oddball task (Figure 9A) alcohol induced changes to SPN amplitude (illustrated with upward arrow in the ERP inset) correlated with participant’s self-reported drinking behaviour (AUDIT) (r = -0.59, p = .002). Alcohol reduced the SPN in participants who drink less, but increased the SPN in participants who drink more. The same trend was present, but non-significant, in the Regularity task (r = -19, p = .347). In the Oddball task, alcohol induced subjective light-headedness correlated with alcohol induced P300 reduction. The direction of this effect is surprising: Participants who were more subjectively affected by alcohol bucked the overall trend for alcohol to reduce P300. However, this was driven by two anomalous participants, with usually high alcohol induced light-headedness.
In the Regularity Task (Figure 9B), alcohol induced change in behavioural performance negatively correlated with alcohol induced change in N1 drop. Also in the Regularity task, alcohol induced changes in behavioural performance correlated with alcohol induced changes in subjective alertness.
We should emphasize that the variables in the correlation analyses were differences between alcohol and placebo conditions. We can also run the same analysis with amplitude averaged across alcohol and placebo conditions. Although the literature suggests N1 and P300 components are altered by chronic drinking, there was no evidence that amplitude of any of our ERP components correlated with AUDIT, and no other correlations were significant at the 0.01 level (see supplementary materials on SPN catalogue /Project 43).