The study conducted a temporal bisection task to investigate the effect of crowding on time perception, independent of valence and arousal. To eliminate their impact, we carefully chose three types of pictures: non-crowding, crowded objects, and crowded people, and observed no significant differences in valence and arousal scores among these picture types (Fig. 1). The results of Experiment 1 indicated that crowding did not affect time perception significantly on sub-second timescales. However, Experiment 2 demonstrated that crowding modulated time perception on supra-second timescales. Then we employed a linear mixed model (LMM) to examine the prediction of crowding, valence, and arousal to the bisection point (BP). The LMM revealed a significant linear relationship between crowding and BP, while valence and arousal failed to predict BP. Finally, we used dominance analysis to reveal the relative importance of crowding, valence, and arousal in predicting the BP, and found that crowding completely outperformed valence and arousal, with crowding explaining more than 91% of overall average additional contributions while valence and arousal together explaining less than 9%. In summary, the findings provide evidence that crowding significantly modulates time perception on supra-second timescales, even when valence and arousal are controlled for three types of pictures.
Crowding and motivation are closely intertwined. High crowd density can be a source of mental stress and too much information can lead to a negative mood state (Schmidt & Keating, 1979). For instance, in crowded public transport, passengers’ discomfort can arise from standing instead of being seated, less opportunities to use time during the journey, and the physical closeness of other travelers (Haywood et al., 2017). This can lead to negative emotions and stress (Bruins & Barber, 2000; Evans & Wener, 2007). Furthermore, crowding can also reduce an individual’s sense of autonomy, increase the sense of personal space invasion (Lawrence & Andrews, 2004; Maeng et al., 2013; Maeng & Tanner, 2013; Nieuwenhuijsen & de Waal, 1982), and reduce individuals’ freedom of activity and control over the environment (Consiglio et al., 2018; Rompay et al., 2008; Stokols, 1972). Therefore, individuals often hope to end the crowded travel and escape from the unpleasant environment as soon as possible (Sadeghi et al., 2023a), which is known as the withdrawal motivation of individuals in the crowded environment (Maeng et al., 2013; Maeng & Tanner, 2013).
We found that crowding can modulate time perception, independent of the valence and arousal (Fig. 5 and Table 1). Given that the crowding is accompanied by the withdrawal motivation (Maeng et al., 2013; Maeng & Tanner, 2013), out results are consistent with the motivation dimension model of time perception, in which withdrawal motivation should slow down the subjective time, and the motivation directly modulates time perception, rather than being mediated by affective valence and arousal (Gable et al., 2016, 2022; Gable & Poole, 2012). Supporting for the motivation dimension model of time perception, Yin et al. (2021) reported that approach and withdrawal motivations modulate time perception after controlling for valence and arousal of emotional images. More evidence suggests that emotion and motivation induce attention biases (Cisler & Koster, 2010) and further modulate time perception (J. Liu & Li, 2020; Yin et al., 2023). Electrophysiological evidence has been obtained to support that emotion modulates time perception through the attention system (Tamm et al., 2014; Vallet et al., 2019). Especially, Yin et al. (2021) employed contingent negative variation (CNV) to examine the processing mechanism of different motivations affecting time perception. CNV is a well-known event-related potential (ERP) component that has been shown to be associated with temporal encoding (Wiener et al., 2012), and more attention assigned to temporal information leads to a longer perceived duration and a larger CNV (Chen et al., 2007; Y. Liu et al., 2013). Angry expressions are negative stimuli with approach motivation, and fearful expressions are also negative stimuli but with withdrawal motivation. According to the attentional perspective of temporal processing (Coull et al., 2004; Macar et al., 1994; Zakay & Block, 1997), approach motivation attract more attention than withdrawal motivation, less attention was paid on time in the angry condition than that in the fear condition, therefore, the perceived time of angry expression is shorter than that of fear expression, and the amplitude of CNV induced by angry expression is lower than that of fear expression (Yin, Cui, et al., 2021). Therefore, compared with uncrowded conditions, withdrawal motivation allows participants allocate more attention to time and less attention to crowded images, resulting in longer perceptual time in crowded conditions.
According to the theory of magnitude (ATOM), time, space, and quantity are all part of a generalized magnitude system, which can lead to cross-dimension interference; the parietal cortex is believed to be the locus of this magnitude system (Bueti & Walsh, 2009; Walsh, 2003a). The parietal cortex is a region of the brain that has been implicated in working memory (Chai et al., 2018). For instance, Jonides et al. (1998) found that parietal regions are part of a network of brain areas that mediate the short-term storage and retrieval of phonologically coded verbal material. Koenigs et al. (2009) also found that the superior parietal cortex is critical for the manipulation of information in working memory. Cui et al. (2022) used event-related potentials (ERPs) to identify the neural basis of cross-dimension interference. They found that the parietal P2 and P3b component index updates a common magnitude representation of spatiotemporal information in working memory, and the neural source of P2 and P3b was located in the parietal cortex. Based on the above theoretical considerations and empirical evidence, it can be inferred that the cross-dimension interference of quantity on time may occur in working memory.
We found that crowding modulated time perception on the supra-second timescales rather than the sub-second timescales. An automatic timing system measures sub-second intervals without attentional modulation, and a cognitively controlled timing system is more involved in the measurement of supra-second intervals with attention and working memory (Lewis & Miall, 2003b, 2003a, 2006). Behavioral evidence also supports the idea that processing shorter intervals depends on sensory or automatic processing while processing longer intervals requires cognitive resources (Hellström & Rammsayer, 2004; Rammsayer, 2006). Taken together, these findings suggest that the cognitive mechanisms underlying time perception are different for sub-second and supra-second intervals. As previously mentioned, after controlling for emotional valence and arousal, crowding may modulate time perception through withdrawal motivation and cross-dimension interference. The impact of withdrawal motivation on time perception may involve the allocation of attention between temporal and non-temporal information in the attention system (Coull et al., 2004; Macar et al., 1994; Yin, Cui, et al., 2021; Yin et al., 2023; Zakay & Block, 1997); the impact of quantitative information on time involves the mutual interference of different dimensional information in a generalized magnitude system located in parietal cortex (Bueti & Walsh, 2009; Walsh, 2003a), which is has been implicated in working memory (Chai et al., 2018). Therefore, the finding that crowding modulates time perception on supra-second timescales rather than the sub-second timescales is consistent with the theory of automatic and cognitively controlled timing systems.
This study has the following advantages and limitations. Firstly, the study effectively controlled for the impacts of valence and arousal on time perception. We carefully selected three types of crowding pictures so that there was no significant difference in valence and arousal scores among the three crowding types. Then, we used LMM and dominance analysis to confirm that valence and arousal did not affect subjective time. The study provided strong evidence that crowding can modulate time perception even after valence and arousal were excluded. Secondly, the study examined the influence of crowding on sub-second and supra-second timescales, respectively, based on the theory of automatic and cognitively controlled timing systems. We found that crowding modulated time perception on the supra-second timescales rather than the sub-second, which helps to establish a connection between crowding and existing theories in the psychology of time. However, one limitation of this study is that it does not distinguish the influence of withdrawal motivation on time perception from that of quantity. Crowding is accompanied by withdrawal motivation and quantity, but the mechanisms of their impacts on time are different. The withdrawal motivation modulates time through attention, and the cross-dimensional interference of quantity on time may occur in working memory. Further research on these two mechanisms in crowded situations can not only provide answers to how withdrawal motivation and magnitude processing affect time perception but also deepen our understanding of the interaction between cognition and motivation. Future studies could combine behavioral experiments with cognitive neuroscience techniques to separate the mechanism of withdrawal motivation and quantity in the modulation of time perception in a crowded environment.
In summary, this study controlled the valence and arousal of three types of crowding pictures, and investigate the effect of crowding on time perception on sub-second and supra-second timescales. We found that crowding modulated time perception on the supra-second rather than the sub-second timescales. The LMM and dominance analysis verified that valence and arousal did not significantly modulate time perception. Crowding includes withdrawal motivation and quantity, which are considered to be influencing factors of time perception after excluding valence and arousal. Withdrawal motivation modulates time perception through attention system, and the cross-dimension interference of quantity on time may occur in working memory. These explanations are consistent with the idea that automatic timing systems measure sub-second time intervals without attentional modulation, while cognitively controlled timing systems are more involved in supra-second interval measurements with attention and working memory.