نقش انگیختگی فیزیولوژیک در ادراک زمان: شواهد فیزیولوژیکی روانی از یک الگوی تنظیم احساسات

Time perception, crucial for adaptive behavior, has been shown to be altered by emotion. An arousal-dependent mechanism is proposed to account for such an effect. Yet, physiological measure of arousal related with emotional timing is still lacking. We addressed this question using skin conductance response (SCR) in an emotion regulation paradigm. Nineteen participants estimated durations of neutral and negative sounds by comparing them to a previously memorized duration. Instructions were given to attend either to temporal or to emotional stimulus features. Attending to emotion with negative sounds generated longer subjective duration and greater physiological arousal than attending to time. However, a shared-attention condition showed discrepancy between behavioral and physiological results. Supporting the idea of a link between autonomic arousal and subjective duration, our results however suggest that this relation is not as direct as was expected. Results are discussed within recent model linking time perception, emotion and awareness.

Time perception is crucial in numerous everyday life activities and, more generally, in the generation of adaptive responses to the environment. Various studies have shown that humans and other animals are able to estimate time accurately in the millisecond to minutes range. This ability is however highly context dependent and recent studies have pointed out the influence of an emotional context on time judgment (Angrilli et al., 1997, Droit-Volet et al., 2004, Noulhiane et al., 2007 and Tipples, 2008). More specifically, emotional events are consistently reported to last longer than neutral ones. Such distortion of time is thought to be linked to fluctuations of the physiological arousal level (Droit-Volet et al., 2004, Droit-Volet and Meck, 2007 and Noulhiane et al., 2007). Yet, no study directly tested this assumption. The present study aimed at exploring the relation between physiological arousal and the emotional effect on time judgment by (1) introducing a measure of autonomic arousal and (2) modulating arousal by cognitive control of emotion. Dominant models of time perception assume that our ability to perceive time relies upon a pacemaker–accumulator clock, where an oscillator – or pacemaker – produces a series of pulses – i.e. temporal units – and the number of pulses recorded over a given timespan represents experienced duration (e.g. Triesman, Faulkner, Naish, & Brogan, 1990). According to such models, two mechanisms can affect the functioning of the timer: attention and arousal. A large number of studies using dual tasks or distracting paradigms have shown that when attention was distracted from the processing of time, duration was perceived as shorter (Brown, 1997 and Macar et al., 1994). Attentional theories of time perception predict that less attention to time will result in fewer accumulated pulses (e.g. Zakay, 2005). By contrast, arousal is thought to increase the speed of the pacemaker, resulting in a longer perceived duration (e.g. Burle & Casini, 2001). Only a few studies have focused on the question of how arousal induced by emotion influences the processing of time. Using a temporal bisection task, Droit-Volet and collaborators (2004) showed that the perceived duration of faces expressing emotion was longer than for those with a neutral expression. Replicating this finding, Tipples (2008) showed that this was particularly true of faces expressing anger. In a study using standardized emotional material, we manipulated positive and negative sounds lasting 2–6 s assessed in terms of arousal, in a reproduction and a verbal estimation task (Noulhiane et al., 2007). Results showed that emotional sounds were perceived as longer than neutral ones. Interestingly, the emotional effect was greater with negative stimuli (Noulhiane et al., 2007), which are thought to induce stronger physiological and cognitive responses than neutral or positive stimuli (Caccioppo & Gardner, 1999). Overall these studies thus support activation-based models of time perception postulating that a higher level of arousal generates longer subjective durations. However, while physiological activation is assumed to be responsible for the lengthening of experienced duration, none of these studies used physiological measures able to assess changes in activation level. Electrodermal activity is an indirect measure of the sympathetic autonomic activity frequently used to index modulations in global physiological arousal level (e.g. Boucsein, 1992). More specifically, skin conductance response (SCR) is widely used in the study of emotion and is thought to vary with the level of arousal induced by emotional stimuli (Kreibig, 2010). This measure thus seems to be a valid, non invasive method, to account for the assumed role of physiological arousal in the lengthening effect of emotion on time perception. With the aim to manipulate variation of arousal, we chose to study time perception in the framework of emotion regulation. Some studies have pointed out the key role of arousal – and not valence – in the conscious regulation of negative emotion (e.g. Dillon & LaBar, 2005). Such proposal leads to the idea that the cognitive control of emotion might constitute a good way to observe variation of physiological arousal and its related effect on timing. Several means of investigating emotion regulation exist and the most commonly used consists in asking participants to voluntary enhance or decrease emotion generated by a stimulus (e.g. Beauregard et al., 2001, Kim and Hamann, 2007, Ochsner and Gross, 2005, Ochsner et al., 2004 and Ohira et al., 2006). Another way to study emotion regulation is to manipulate attention to versus away from emotion. According to Gross and collaborators ( Gross, 1998b; Gross & Thomson, 2007), this strategy, termed ‘attentional deployment’, have an impact early in the emotion-generative process. In that, it should ensure a good manipulation of top-down control of arousal induced by emotion and its related effect on time perception. This hypothesis was tested in a duration comparison task, including neutral and negative sounds of varying emotional intensity. The choice of using negative stimuli was guided by the fact that negative events are thought to generate stronger physiological responses than positive events do (e.g. Cacciopo & Gardner, 1999). In a previous experiment, we showed that the effect of emotion on time judgment was maximal with short duration stimuli and progressively declined with the stimulus duration (Noulhiane et al., 2007). Therefore, we chose to focus on short duration processing – i.e. 2 s – in order to maximize the emotional effect on time perception. Attentional deployment was manipulated by prior instructions to attend either to time or to emotion or to both dimensions. These instructions may be linked to the three traditional conditions used in emotion regulation paradigms, i.e. increasing, decreasing and maintaining emotion. Attending to emotion is expected to enhance emotion, while attending to time is expected to down-regulate emotion. Attending to both dimensions may reflect an intermediate condition, which could be assimilated to the ‘maintaining emotion’ condition. We expected to replicate the relative overestimation of emotional compared to neutral sounds when participants directed their attention toward emotional intensity or divided their attention between time and emotion, but not when they had to focus their attention on time and ignore emotional features. Furthermore, we expected that longer time estimates will be associated to higher levels of SCR.

Integrating an objective measure of arousal in the study of emotional timing, the present experiment accredits prior assumptions that the modulation of physiological arousal is related to variation of subjective time (e.g. Burle & Casini, 2001). It nonetheless suggests that this relation is not as direct as suggested by initial theories of internal clock (Treisman, 1963). Rather heightened arousal and subjective dilation of time are likely to be concomitant of a same phenomenon. In line with recent findings attempting to link time perception to self-referential processing, we propose that enhancement of self-awareness – induced by high salience – constitutes the basis for time dilation in an emotional context.