Interoceptive signals and emotional states shape temporal perception through heart rate modulation

Abstract

Background: This study investigated the relationships between interoceptive signals, emotions, and time perception, with a particular focus on the mediating role of heart rate (HR). Emotional video stimuli were used to elicit specific emotional responses, while participants' attentional focus was manipulated (internal vs. external) to examine its influence on temporal judgments.

Methods: We tested several hypotheses using a combination of linear mixed models and Bayesian mediation analysis. Emotional content, heart rate, attentional focus, and interoceptive accuracy were analyzed for their effects on time perception. Participants viewed 36 video clips designed to elicit positive, negative, or neutral emotions, and their duration estimation errors, normalized heart rates, and subjective ratings were recorded.

Results: Emotional content significantly influenced time perception. Negative and neutral videos were underestimated in duration, while positive videos showed smaller estimation errors. Heart rate partially mediated the effect of emotions on time perception, with slower heart rates linked to greater underestimation of durations. However, this mediating effect was smaller compared to other potential mechanisms not explored in this study. Contrary to prior research, no correlation was found between interoceptive accuracy and time perception precision, likely due to methodological differences in task design and measures of interoception. An internal focus of attention was associated with greater underestimation of time and lower normalized heart rate. However, no significant interaction was observed between attentional focus, heart rate, and emotional valence.

Conclusion: These findings highlight the multifaceted nature of time perception, emphasizing the role of both physiological processes and subconscious interoceptive signals. The partial mediation of heart rate underscores its importance in shaping temporal judgments, while the lack of interaction with attentional focus suggests that these effects may be driven by unconscious mechanisms. The results contribute to a deeper understanding of how emotions and bodily signals interact to shape time perception and underscore the need for further research into individual differences and unconscious influences on temporal judgments.

Keywords: awareness; emotions; heart rate; interoception; time perception.

Figure 1

The design of the experiment. The sequence of internal and external attention focus varied among participants.

Figure 2

Error of video duration estimation when viewing negative, neutral, and positive videos. The box indicates the interquartile range (IQR), with the lower edge representing the 1st quartile (Q1) and the upper edge representing the 3rd quartile (Q3). The box contains the middle 50% of the data and medians, while the whiskers extend to the minimum and maximum values. Gray unfilled circles indicate individual data points. The indicated numbers represent the mean values for each group. The p-values for the post-hoc Tukey test are provided.

Figure 3

HR (z-score) when viewing negative, neutral, and positive videos. The box indicates the interquartile range (IQR), with the lower edge representing the 1st quartile (Q1) and the upper edge representing the 3rd quartile (Q3). The box contains the middle 50% of the data and medians, while the whiskers extend to the minimum and maximum values. Gray unfilled circles indicate individual data points. The indicated numbers represent the mean values for each group. The p-values for the post-hoc Tukey test are provided.

Figure 4

Correlation between the mean normalized heart rate of all subjects and the mean pleasantness scores of the video, r = 0.37, p = 0.014. One point corresponds to one video stimulus.

Figure 5

Individual relationships between heart rate and estimation error. (A) Scatterplot of z-scored heart rate (HR) against z-scored time estimation error, with individual participants shown in different colors. Each dot represents one trial. The black line indicates the group-level linear trend (β = −0.24), reflecting the overall association between HR and estimation error across participants. (B) Distribution of individual regression slopes from the relationship between heart rate (z-scored HR) and time estimation error across trials. Each slope represents the estimated change in estimation error per unit increase in z-scored HR for a single participant. The dashed red line indicates the median slope across all participants, while the dotted black line marks zero (no association). A predominance of negative slopes suggests that elevated heart rate was generally associated with underestimation of video duration.