When asked to judge the duration of a face people typically overestimate the duration of angry compared to neutral faces. A novel feature of the current research was the inclusion of secondary manipulations designed to distort timing performance namely, the effects of visual cues (Experiment 1) and action preparedness (Experiment 2). Furthermore, to establish whether the effects are multiplicative with duration, the effects were examined across 2 duration ranges (200 to 800 ms and 400 to 1600 ms). Visual cues and instructions to prepare to act increased the tendency to judge faces as lasting longer. Experiment 1 revealed an unexpected underestimation effect for angry faces presented for short durations (200 to 800 ms). However, the effect was not replicated in Experiment 2 where the results were generally consistent with either an increase the speed of a pacemaker mechanism that resides within an internal clock (Gibbon, Church, & Meck, 1984; Treisman, Faulkner, Naish, & Brogan, 1990; Treisman, 1963) or the widening of an attentional gate (Zakay & Block, 1996) – the temporal overestimation effect for angry faces grew in magnitude from the short to long duration. Experiment 2 also showed that the temporal overestimation for angry faces was reduced in magnitude when participants were asked to prepare to either push or pull a joystick.

How does emotion change the way we perceive time? Studies have shown that we overestimate the duration of faces that express anger of fear–an effect that has been explained as due the speeding of a pacemaker that resides within an internal clock. Here, we test the idea that attending longer to facial threat leads to an overestimation of time. Seventy participants (16 male) estimated the duration of angry, fearful and neutral expressions under conditions designed to either reduce attention to time (by emphasising speedy responses) or lengthen attention to time (by emphasising accuracy). Results were modelled using Bayesian Multilevel Logistic Regression. The results replicate previous findings: speed emphasis reduced temporal sensitivity and led to both a higher overall proportion of long responses and faster reaction times. Facial threat attenuated the drop in temporal sensitivity due to speed instructions supporting the idea that people prolong attention to threat (even when they are not directly instructed to do so). We relate the findings to research into attention bias to threat and more broadly to models of perceptual decision making.

Time perception is malleable ‒ it can be made to speed up and slow down by various experimental manipulations including the presentation of a sequence of auditory clicks and also angry facial expressions. Recent evidence supports the idea that auditory click trains increase accumulation of evidence across time. Here, we test this idea for both angry expressions and auditory clicks by modelling response times (and choice responses) using Bayesian Hierarchical Drift Diffusion Modelling. Two separate groups of participants (Experiment 1; n = 29; Experiment 2; n = 38) judged the duration of angry and neutral facial expressions preceded by either a 3-s sequence of auditory clicks or silence. In both experiments, standard psychophysical analyses showed that both clicks and angry expressions lengthened the perception of time. The original finding came from the analyses of the Drift Diffusion Modelling parameter that represents the speed of information accumulation ‒ the drift rate parameter. Drift rates grew in magnitude with the duration of the face and moreover this effect was larger when the faces were either preceded by clicks or appeared angry ‒ evidence for accelerating temporal accumulation. This novel insight would not have been possible from traditional psychophysical analyses and therefore, the results highlight the potential value of Bayesian Hierarchical Drift Diffusion Modelling as a tool for understanding how we perceive time.

No abstract