Timehas always been one of the most mysterious and challenging aspects of life tofigure out. From physics to philosophy, we are still trying to define theactual existence and meaning of time in our lives. Hard to be conceptualized,it has been a great enigma for centuries now, rising numerous questions. Humans were aware ofdifferent type of times long before they had even formulated the concept oftime itself (Roeckelein, 2008, p. 4). The first attempts to define and studysystematically- albeit naively and non-scientifically- the concept of timesprang from the Ancient Greeks as early as 5th century B.
C. (Hyland, 1994). TheGreek philosopher Aristotle was the first to ask from a psychological point ofview how we perceive time (Nichols, 1989). For Aristotle, time was a directsense perception that derived from the sentient faculty or soul as an immediatefunction under the presentation of primary sensation or of memory and asNichols (1989, p.458) states, Aristotle’s perception of time was a remarkablefirst exposition of time where its essential features have survived in allages.
Centuries later,psychology has put much effort to describe and define time with differentterms. As far as cognitive and experimental psychology is concerned, manyresearchers were interested in exploring the field of time perception (Wearden,2016, p. 5-25).
In the last century, Francois (1927) and Hoagland (1933) werethe first researchers to point out that our time judgments vary as a functionof internal and external contexts. This realization concluded in the idea ofsubjective time, which is referring to our personal evaluation of time,regardless the faithful reading of objective time, e.g., from a counter (Gil& Droit-Volet, 2012). About thirty years later, in the early 1960s, one ofthe today’s most known theories was proposed by the works of Creelman (1962)and Treisman (1963), describing a model regarding a clock-like mechanism whichwas established as “the internal-clock model”(Wearden, 2016, p.19-25).
It waslater on developed into the scalar timing theory, or also referred to as thescalar expectancy theory (SET; Gibbon, 1977; Gibbon, Church & Meck, 1984;Church, 1984), which proposesthat duration discrimination and perception of time are results of an internalclock, memory stores and a decision mechanism. The internal clock is composedof three components: a pacemaker, a switch, and an accumulator. The pacemaker emitspulses, which are sent to the accumulator through the switch. The switch hasthe role of the pulses’ guard. This means that it is the one to control howmany pulses pass through by signaling the start and the end of the time of theevent. Then, the pulses proceed to the accumulator so as to be counted and thenconverted into the perceived duration (Li & Yuen, 2015).
A revisedtheoretical framework for discussing time estimation based on the scalar-timingmodel and on Treisman’s model was presented by Thomas and Weaver (1975), andexpanded by Zakay and Block (1995). They introduced attentional models of timeperception, which in more detail concern information-processing (IP) mechanisms(Buhusi & Meck, 2005), such as the Attentional-Gate Model (AGM; Zakay &Block, 1995). This model proposes that organisms distribute attentionalresources, which are drawn from the same pool, between all performing tasks andthat temporal processing in order to function optimally requires the most of attentionalresources. Consequently, when a non-temporal task is in need of more attention,then fewer resources are available for the temporal one, and thus temporalperformance is impaired (Brule & Casini, 2001).
As a result, fewer pulsesare accumulated and the duration is perceived as shorter. This hypotheticalcognitive model explains why durations seem shorter when more attentionalresources are allocated to non-temporal tasks (Brule & Casini, 2001; Zakay,Nitzan, & Glicksohn, 1983).Based on thesetheories, recent studies acknowledge that time perception can be dramaticallyaffected by variations in external stimulation and by the cognitive state ofthe individual (Droit-Volet & Meck, 2007). Hence, the present research’spurpose is to study how time estimation is being altered by specific featuresthat may capture an organism’s attention. For that reason, we selected humanfaces, which are largely known to attract human’s attention, as well as becapable of attention capture (Langton et al., 2008) even if they aretask-irrelevant (Simpson et al.
, 2014). Preeminently, as it has been supportedby many researches, in comparison with neutral faces, faces that involve facialmovements, such as emotional expressions or gaze, tend to capture attentioneven more (e.g., Vuilleumier, 2002). More specifically,a great number of studies have confirmed that people are highly sensitive togaze direction, an ability that emerges in the very early stages of childhood.Young infants prefer to look at the eyes more than at other regions of the face(Morton & Johnson, 1991) and by the age of 4 months can discriminatestaring from averted eyes (Vecera & Johnson, 1995). Furthermore, there is evidenceof an advantage on the perception of direct gaze that extends to the perceptionof change in direct gaze implying that directly gazing at someone capturesattention more (Yokoyamaô et al.
, 2011). Eye contact and mutual gaze caninfluence face processing, even for a single exposure of an unfamiliar face(Vuilleymier et al., 2005) and facilitate understanding other persons, i.e.
,accessing relevant material in semantic memory (Macrae et al., 2002). As Emery(2000) states from an evolutionary point of view, many species seem to perceiveeye gaze as a crucial social signal and additionally visual processing may havebecome of vital significance due to its attribution in confrontingmorphological, environmental and habitat changes throughout primate evolution.Consequently, eyegaze could be argued to play a significant role in capturing human attentionand as a result influence time estimation into perceiving shorter durationsthan the actual duration presented. Accordingly, there are some supportiveevidence advocating interval underestimation for short durations, such as 1200and 2400 ms, when there was stimulus of direct gaze in comparison with avertedgaze stimulus (Rouchitsas, 2015), as it could have been predicted by theattentional-gate model. However, there are studies that yield opposite results,where the direct gaze stimulus produced longer subjective duration responses(Thones & Hecht, 2016). Notably, both the above-mentioned studies provideonly weak and task-dependent evidence for the influence of mutual gaze onsubjective time.
Nevertheless, it is clear and fair to note that direct gazemay indeed play a crucial role in changing the subjective duration of atemporal interval that is present on the order of several seconds.Even thoughstudies on time estimation and gaze direction are limited, much research hasbeen made concerning the interplay of emotional expression and mutual oraverted gaze. The way in which gaze direction influences emotion perceptionactually depends on the specific type of emotion and the overall social contextin which it is placed. Adams and Kleck (2005) found that for some emotions,such as joy and anger, direct gaze enhances perception, but for other emotions,e.g.
, fear and sadness, averted gaze does. They assumed that this kind ofdifferentiation in enhancement happens because anger and fear indicate thesource of threat (i.e., as part of an early warning mechanism), whereas joy andsadness may just be a social signal indicating a tendency for socialengagement. Additionally, inanother study angry faces were recognized as expressing more anger and that theunderlying emotion was judged as being more intense with a direct than with anaverted gaze, whereas the opposite effect was revealed for the emotion of fear (Sanderet al., 2007).
Looking directly towards someone else can also elicit morepositive and arousing reactions than when looking away, perhaps due to theperception of direct gaze as an affiliative signal (Hietanen et al., 2018;2008). It has been shown that direct gaze is a sign indicating the other personas trustworthy and truthful (Bayliss & Tipper, 2006) and it is also used infunctions such as the regulation of turn-taking in conversation, expressingintimacy, and exercising social control (for a review, see Langton, Watt, , 2000). Both emotional expression and gaze behavior seem to influenceattention and communicate basic behavioral motivations to social approach oravoidance.In theemotion-related domain, many studies have been executed appointing the hugeeffect of emotion, in general, and of facial expressions, in particular, ontime estimation (for reviews, see Droit et al, 2007; Lake, 2016; Lake, Labar, , 2016). However, even though cognitive variables such as emotion have beenshown to influence judgment of duration, this happens not entirelysystematically (Meck et al.
, 2003). Hence, some evidence suggested that notonly attention but also arousal play a crucial role in time perception by influencingdifferent parts of the internal clock (Gil & Droit-Volet, 2011). In moredetail, it has been proposed that arousal influences the operation of theinternal pacemaker by speeding up its rate of emitting pulses and subsequentlythis increases the number of stored units in the accumulator (Gil & Droit-Volet,2012). Consequently, the duration of emotionally loaded stimuli (e.
g., facialexpressions) are being overestimated relative to neutral stimuli. Angrilli etal. (1997) conducted a study in order to investigate the effects of arousal andvalence on time perception. They found that there was no main effect of arousalor valence, but a significant interaction between those two variables. Morespecifically, at low levels of arousal the duration of negative slides wasunderestimated, while the duration of positive slides was overestimated.
Incontrast, at high levels of arousal, positive slide duration was underestimatedrelative to the duration of negative slides. A great deal of data has beencorroborative to this arousal-driven temporal distortions by indicating alengthening effect produced by the perception of emotional faces and especiallyof angry expressions, which are considered as more arousing due to their life-threatening meaning (Doi & Shinohara, 2009; Droit-Volet, Brunot, , 2004; Fayolle & Droit-Volet, 2014; Fayollea, Gil -Voleta, 2015; Gil et al., 2011; Gil & Droit- Volet, 2012; Li et al.,2009)On the other hand,the advocates of the attentional gate model suggest that time intervals shouldbe underestimated when emotional stimuli are present due to the fact thatattention is allocated from time estimation to the cognitive process of vitalimportance signals (Brule et al., 2001; Zakay et al., 1983). Many studies havehighlighted through different types of examination (e.g.
, detection tasks orStroop-like paradigm) an automatic processing of emotional expression and aspontaneous bias to orient attention towards threatening cues, e.g., angryfaces (Vuilleumier, 2002).
Moreover, threatening stimuli not only captureattention but also delay the disengagement of attention from it (Belopolsky,Devue, & Theeuwes, 2011).Lui and her colleagues (2011) found supportiveevidence that perceived time was shorter when it was preceded by an emotionalas compared to a neutral distractor indicating that emotional experiences maydecrease temporal estimates. However, the most contemporary data propose acombined mechanism, which involves both attention and arousal (see Figure 1).Arousal seems to influence mostly the pacemaker rate and as Angrilli et al.(1997) implicated this phenomenon occurs only for very short durations, lessthan 1s, whereas changes in attention, as reported earlier, seem to affect theactivity of a switch or gate that influences the number of pulses collected inthe accumulator but for longer durations. Figure 1. An oversimplified way of demonstrating the derivedof emotion mechanisms of arousal and attention, as modulatory factors on timeestimation.AV1 The aim of thepresent study is to investigate and describe the temporal distortions insubjective time, which occur as a result of gaze behavior and facial emotionalexpressions.
Two types of gaze were examined, direct gaze and averted to theleft gaze, as well as three facial expressions, anger, fear, and happinessalong with a neutral expression as baseline. Many previous studies presented aserious limitation as far as the type of stimulus is concerned. They usedmostly static images (Angrilli et al., 1997; Bar Haim et al.
, 2010; Doi et al.,2009; Droit- Volet et al., 2004; Droit- Volet et al.
, 2015a; Droit-Volet etal., 2015b; Gil et al., 2011; Gil et al., 2012), which cannot fully reflectreal life situations. Under those circumstances, various problems emergeconcerning ecological validity because facial behavior is a combination ofdynamic information rather than static patterns of portrayed emotionalexpressions (Krumhuber, Kappas, & Manstead, 2013). Hence, it is veryimportant the experimental variables to be to the greatest extent closer toreality including all the dynamic aspects facial activation can provide, as anattempt to imitate the complexity of actual social interactions of dailyexperiences. Only a few recent studies have focused on the different outcomesproduced by morphing photographs and static displays of emotions on timeestimation, concluding in a possible enhancement of the lengthening effect whenthe stimulus was to some extent more dynamic (Fayolle et al.
, 2014; Li et al.,2015). Although the above-mentioned researches offered some interestingresults, it must be stated that the utilization of morphing photographs in grayscalestill undermines considerably the attempt to approach real-life perceived facesdue to the fact that morphing has been shown to result in less accurate effects(Krumhuber et al., 2013).
Conversely, our experiment contained validated coloredvideos from trained actors (Van der Schalk, Hawk, Fischer, & Doosje, 2011),in order to emulate as much as possible real experiences. Hence, our experimentcould be considered as original in attempting to research authentic dynamicaspects of emotional expressions and gaze behavior on the perception oftemporal intervals. Another essentialfeature under consideration in studies of time perception is the temporalprocedure to be followed.
Each estimation method varies in the use of differentcognitive processes and as a result it has been stated that different approachesproduce different outcomes (e.g., Gil et al., 2011). Henceforth, it is crucialto choose if the duration judgments are prospective or retrospective as well asto specify the exact method of examining interval timing.
In our study, we usedthe prospective paradigm, which refers to a situation in which a person isaware, during a time period that he or she needs to estimate this duration. Ithas been argued that prospective duration timing depends on attention demandingprocesses (Zakay & Block, 2004). Accordingly, since our aim is to examinethe possible effects of attention on time estimation, prospective paradigmappears to be the most appropriate method to be followed., One of the mostapplied methods in literature of perception and processing of temporalinformation in the field of prospective timing, especially regarding theeffects of facial expressions, is the temporal bisection paradigm (for areview, see Kopec & Brody, 2010).As a temporal discrimination task,participants are required to compare temporal stimuli to two reference stimuli,”long” and ”short”, held in memory.
It is a well-known and largely testedtemporal task which has though often yield mixed and ambiguous results (Wearden,2016, p.72-73). For this reason, we selected this method to examine timingjudgments in our experiment in order to provide an additional insight for thiscontroversial task when investigating dynamic stimuli. Furthermore, it seemslike a logical application since our tested population involves only participantsof age which in a meta-analysis by Kopec and Brody (2010), has been shown thatindividuals performing the bisection task should be only over 8 years old.In summary, ourmain goal in the present experiment is to study the interplay of emotion andgaze on time estimation by the use of dynamic stimuli.
If the arousal- drivenmechanism is correct, then we shall find supportive evidence to currentbibliography concerning emotion and overestimated time judgments. On the otherhand, if AGM models are accurate, it is our strong belief that both emotion anddirect gaze will produce significantly shorter subjective judgments. AV1usea better legend this just shows the AGM and where arousal or attention isexpected to affect….and also provide the reference where you took this and madeyour own graphic