Abstract:
In this work, we analysed the effect of a 2-hour period of sleep or wakefulness on behavioural and cardiovascular responses during aversive conditioning. Specifically, we examined reactions of heart rate (HR), heart rate variability (HRV) and pulse wave amplitude (PWA), measured through photoplethysmography. We also examined the change in behavioural measures in response to each stimulus as well as emotional state and vigilance before and after each test. Additionally, we calculated correlations between the conditioned responses and stable personality traits and tested whether sleep quality was affected by the conditioning experiment.
A total of 18 participants took part in four identical series of tests on two afternoons. One series of tests was performed before (pre-intervention) and one after (post-intervention) a 2-hour period of sleep or wakefulness. Participants were asked to sit still and attentively listen to 5 different tones, presented via headphones. Aversive conditioning was performed by using a very loud aversive tone (unconditioned stimulus, US+). This US+ was paired with an initially emotionally neutral tone (conditioned stimulus, CS+). Furthermore, there was a neutral standard tone as well as two other neutral tones (US- and CS-) which were also paired with each other. Each series of tests consisted of three blocks. In each of these blocks, 280 standard tones, 60 CS+ and 60 CS- were presented. In between blocks there was a short break and emotional responses towards each stimulus were assessed. In the first block the aversive tone was not introduced, thus the CS+ was unpaired (0% reinforcement rate). In the second block every second CS+ was followed by a US+ (50% reinforcement rate). In the third block each CS+ was followed by a US+ (100% reinforcement rate). During all tests finger plethysmography was measured continuously. All other results were acquired through questionnaires.
After the period of wakefulness, as compared with the pre-intervention session, HR decreased, whereas HRV increased. These changes were much less pronounced after the 2-hour nap, resulting in a strong and significant interaction between time (pre-intervention versus post-intervention) and the type of intervention (sleep versus wakefulness). This might be due to the circadian dynamics of cardiovascular parameters in the control condition, whereby sleep is supposed to break this circadian trend. Another, less plausible, interpretation might be a general sympathetic surge of the sleep group directly after waking up. Other studies hold an ongoing parasympathetic activation responsible for less pronounced cardiovascular responses following sleep. PWA decreased after the period of wakefulness. In contrast to HR and HRV, possibly its response was not concealed by circadian variance. All three cardiovascular measures significantly responded to aversive stimulation, but PWA, in contrast to HR and HRV, also depended on the rate of the aversive stimulation indicating that PWA provided additional information beyond that of HR and HRV.
Self-assessment manikins (SAM) were used to evaluate emotional responses during the conditioning experiment. Valence and arousal scores in reaction to each stimulus were collected in between test blocks. In line with the previous studies, subjective emotional responses exhibited fast learning, but also fast extinction. The type of intervention (i.e., sleep or wakefulness) did not have an effect on these measures. This is contrary to the cardiovascular measures that were affected by the type of intervention (see above), indicating the lack of a simple relationship between autonomic and subjective parameters.
The State-Trait Anxiety Inventory (STAI) questionnaire and the Positive and Negative Affect Schedule (PANAS) questionnaire were used to measure emotional state measures before and after each of the four test series. Anxiety levels and negative affect increased with conditioning, and the increase was stronger before than after interventions, suggesting a preserved trace of memory after the 2-h interval.
The nap had refreshing effect on participants, manifested in an increase of positive affect, decrease of subjectively experienced sleepiness, and faster reaction times in a vigilance task after the nap compared to wakefulness.
As expected, participants with a higher sensation seeking scores were less aroused by the CS+ than low sensation seeking individuals. No personality scores showed any correlation with cardiovascular measures.
Quality of sleep was not affected by the experiment, but participants needed less sleep after the sleep day than after the control day.
For future studies it would be interesting to analyse the effect of the circadian rhythm on cardiovascular measures, for example by performing a similar experiment with a morning intervention.
Additionally, with longer interstimulus intervals it would be possible to measure cardiovascular reactions towards each individual stimulus. These could be correlated with the emotional reactions towards each stimulus. A continuous measurement of cardiovascular information during the interventions could also help to find an explanation for the significant difference of cardiovascular reactions in the post-intervention wake group compared to the sleep group.
Print-on-Demand