Supplementary Figure 17: HEX impacted self-reported mood in ASD but not TD

A. The influence of HEX on the positive mood rating relative to control of TD (n =16, white) and ASD (n =17, red). Each circle represents the positive mood rating of a participant following HEX (y-axis) and Control (x-axis). B. The positive mood of the ASD participants increased significantly compared with TD (F1,31 = 7.8, p = 0.009). A two-way repeated-measures ANOVA with conditions of Condition (Hexadecanal/Control), Time (Beginning/End of the experiment) and a categorical independent factor of Group (TD/ASD) was applied to each of the mood subscales (Positive/Negative low/Negative high/Sexual arousal). The ANOVA on Positive mood revealed a significant interaction of Condition X Time X Group (F 1,31 = 7.8, p = 0.009, Cohen's d' = 0.97) (Supplementary Figure 19). This reflected a significant increase in the positive mood of the ASD group alone following exposure to undetected HEX (ASD: Δ Hex = 86.3 ± 174.1, Δ Control = −68.5 ± 128.1, t(16) = 4.1, p = 0.0008, TD: Δ Hex = −37.1 ± 136.9, Δ Control = 2.0 ± 174.4, t(15) = 0.65, p = 0.52). ANOVAs on the other subscales revealed no interaction or main effect that involved Condition (all F < 2.5, all p > 0.1). In other words, the influence of HEX on self-reported mood in this study was limited to one subscale alone. Although this again points to dissociable impact of HEX in TD and ASD, we are cautious in interpretation of this result. This is because unlike the psychophysiological results that emerged again and again throughout all experiments in this study, this self-reported mood result is the only significant case throughout the study. Although its extent is such that it in fact survives correction for multiple comparisons following from the above, we nevertheless retain caution in interpretation. All tests were two-tailed, all centers reflect mean, all error bars reflect SEM. ** = p < 0.01.