Fig 1.
Study design Experiment Part 1.
The timeline displays sweat sampling conditions: baseline, fast stress (i.e., SAM activity), slow stress (i.e., HPA axis activity), and measurements: heart rate, cortisol, sweat production. T = time. T+20 = 20 minutes passed since the start of the experiment.
Fig 2.
Physiological outcomes senders.
Physiological measurements obtained from senders during three conditions: pleasant-neutral baseline, preparation for a speech (i.e., fast stress, SAM axis activity), and recovery from speech preparation (i.e., slow stress, HPA axis activity). (A) Mean heart rate (bpm) per condition. (B) Mean sweat production (mg) per condition. (C) Mean salivary cortisol (nmol/l) over time, including a measure following lab entry and measurements before and after each experimental condition (Base = Baseline; Fast = Fast stress; Slow = Slow stress). Error bars ± 68% within-subjects confidence interval (CI) of the main effect (for the formula leading up to the calculation of the CI, see [66]; for the choice of a 68% CI, see [67]).
Fig 3.
Mean facial muscle activity of receivers over time as a function of odor.
(A) Mean corrugator supercilii activity (i.e., brow knit) following odor onset (in seconds). (B) Mean medial frontalis activity (i.e., brow lift) following odor onset (in seconds). Facial muscle activity displayed here was measured before the start of the facial expression classification task, to isolate the effect of odor. Error bars reflect 68% within-subjects CI of the interaction between odor and time.
Fig 4.
Mean facial muscle co-activation of receivers over time as a function of odor.
Facial muscle activity displayed here was measured before the start of the facial expression classification task, to isolate the effect of odor. Above each bar, the time after odor onset (in seconds) is depicted (see Y-axis). The more each bar is located toward the upper-right end point (vs. bottom-left starting point) of the dashed diagonal, the more the medial frontalis and corrugator supercilii muscles co-activated (μV), resembling a fearful facial expression [cf. 11, 12].
Fig 5.
Mean facial muscle activity of receivers per odor condition during classification of presented (emotional) facial expressions.
Odor condition: Baseline, fast stress, slow stress. Facial expressions that had to be classified: Neutral, happy, fear, disgust. For clarification purposes, the display of mean facial muscle activity on the emotional facial expression classification task was collapsed over the variable noise level (20%, 40%, 60%, 80%, 100%). (A) Mean corrugator supercilii activity, averaged over 1 second following the onset of the presented expression. (B) Mean medial frontalis activity, averaged over 1 second following the onset of the presented expression. Error bars reflect 68% within-subjects CI of the main effect of odor.
Fig 6.
Mean speed and accuracy of receivers to classify (degraded) facial expressions per odor condition.
(A) Mean reaction time (ms) of facial expression classification (disgust, fear, happy, neutral) per odor condition (baseline, fast stress, slow stress), collapsed over noise levels (for clarification purposes). (B) Mean accuracy (proportion) of facial expression classification (disgust, fear, happy, neutral) per odor condition (baseline, fast stress, slow stress), and noise level (20–100%). Error bars reflect 68% within-subjects CI of the interaction between odor and facial expression.
Table 1.
Receivers’ ratings of sweat sampled from senders under different conditions.