The potential mechanisms underlying the effect of acute alcohol use on duration perception Authors: Yunpeng Liu, Minghui Lu, Huazhan Yin, Chun Yang, Dehua Wu Published: 2024 ABSTRACT AbstractAcute alcohol consumption has been found to cause duration perception distortions, but the directions of these distortions are not consistent. The mechanisms underlying this effect are also unclear. The present study seeks to elucidate the effect of acute alcohol consumption on duration perception and the mechanisms involved. Forty‐one participants in the placebo group and 40 in the alcohol group completed time bisection tasks, attentional network tests, digit span backward tests and arousal reports to evaluate their duration perception, attentional network, working memory capacity and arousal. The results showed that the alcohol group overestimated duration compared to the placebo group. The alcohol group also showed increased arousal, impaired executive control of attention and reduced working memory capacity. Arousal mediated the effect of acute alcohol consumption on duration perception, whilst working memory capacity masked this effect. The findings are discussed based on the Scalar Timing Model and the Cognitive Resource Allocation Model. INTRODUCTION The distortion of duration perception is a fundamental manifestation of human adaptation to the objective environmen. 1 Many factors have been identified as contributing to altered duration perceptions, including emotional experiences, 2 , 3 cognitive impairments 4 and addictive substances. 5 , 6 Previous research has found that individuals distort their duration perception following acute alcohol use, 7 but the direction of the distortion varies across findings, with some perceiving the duration to be prolonged (overestimating duration, 8 , 9 ), others perceiving it to be shortened (underestimating duration, 10 , 11 ) and still others having relatively normal duration perception. 12 , 13 The mechanism of distortion of duration perception due to acute alcohol use and the cause of variation in the direction of this distortion are currently unknown. Therefore, the present study seeks to investigate the performance and potential mechanisms of acute alcohol use affecting duration perception, which may assist in clarifying the causes of distorted duration perception after alcohol use and deepening the understanding of the effects of addictive substances on duration perception. Scalar Timing Model (STM) is considered to be one of the influential models for explaining human temporal judgement. The model suggests that duration perception is not only dependent on an internal clock, but also encompasses a memory phase and a decision‐making phase. 14 The internal clock phase is where a pacemaker emits pulses at a constant rate, which are accumulated by an accumulator after passing a switch to form a perceptual representation of duration. The possibility of attending to timing information determines the opening or closing of the switch and influences whether or not the pulses are passed through. In the memory phase, the pulses accumulated are stored in working memory and reference memory, corresponding to the current and past perceived durations, respectively. In the decision‐making phase, the amount of currently stored pulses is compared with that stored in the past, thereby enabling the judgement of the duration at the present moment to be completed. Previous research has shown that arousal affects the pulse emission rate of the internal clock phase. 3 Increased arousal can accelerate the pulse emission rate and ultimately result in overestimation of duration. 3 However, both the delayed switch closure due to the failure to capture the timing information in time and inability to continuously attend to the timing information reduce the passage of pulses and ultimately result in duration underestimation. 15 , 16 , 17 , 18 Furthermore, it has also been demonstrated that either the reduction of working memory resources allocated to timing information processing or the reduction of working memory capacity itself reduces individuals' storage of pulses, which results in duration underestimation. 19 , 20 , 21 , 22 Overall, variations in each part of the timing process can distort the perceived duration. Previous studies have demonstrated three types of performance in duration perception after acute alcohol use: overestimation of duration, 8 , 9 underestimation of duration 10 , 11 , 23 , 24 or no significant distortion of duration perception. 12 , 13 , 25 , 26 , 27 , 28 In the presence of duration perceptual distortions, alterations in pulse emission rate have previously been used primarily to reveal observed distortions. For example, the overestimation of duration after acute alcohol use has been attributed to accelerated pulse emission rate by alcohol, 8 whilst the performance of underestimated duration has been due to slowed pulse emission rate by alcohol. 11 The role of pulse firing rate actually mirrors the specific role of arousal. 3 Furthermore, it has been suggested that the impairment of working memory also contributes to the effect of acute alcohol use on duration perception 8 ; the precise role of working memory, as well as attention, remains uncertain. Thus, their role in engaging in the duration perception distortions following acute alcohol use needs to be revealed. Acute alcohol use impairs attention and working memory. 29 , 30 In terms of attention, alcohol impairs executive control, which is reflected in the alcohol's impairment of executive control in the attentional network. 31 Impaired executive control means that individuals are unable to inhibit conflict in favour of the required responses. 32 Consequently, following drinking, individuals' immediate or sustained attention to timing information can be disrupted by unsuppressible information. Moreover, in terms of working memory, previous research has demonstrated that alcohol use significantly impairs individuals' working memory performance in span tasks, implying that alcohol impairs storage processes. 30 Given that storage is reflected in capacity, acute alcohol use thus may result in individuals being unable to maintain the storage of timing information due to impaired working memory capacity. Both the failure to immediate or sustained attend to timing information and the inability to maintain timing information contribute to the underestimation of duration. 17 , 21 Thus, it is reasonable to infer that executive control and working memory capacity function in the effect of acute alcohol use on duration perception. However, since duration after acute alcohol use has been previous found to be either overestimated or underestimated, or even unchanged, executive control and working memory capacity, as well as arousal, may function synergistically in the effect (all three being impaired and contributing to duration underestimation) or antagonistically (increased arousal contributing to duration overestimation, whilst reduced executive control and working memory capacity contributing to duration underestimation), which requires further exploration. The present study aims to investigate the effect of acute alcohol use on duration perception and to explore the role of arousal, attention and working memory in the effect. The time bisection task, the attentional network test and the backward digit span test were used to measure duration perception, attentional functions and working memory capacity, respectively. These measures have been used previously to elucidate the underlying mechanisms of alcohol use disorders affecting duration perception. 21 In addition, the arousal state was measured to reflect the pulse emission rate. It was hypothesised that (1) acute alcohol use could affect duration perception and that (2) this effect was influenced by arousal, executive control and working memory capacity. METHODS Participants A sample size of 74 participants was calculated using a repeated measures analysis of variance (ANOVA) in G*power 3.1, with α set at 0.05, 1–β at 0.8 and an effect size f at 0.25. A total of 81 participants (mean age = 23.67 ± 3.95 years, range 17 to 28 years, 41 males) was recruited from Hunan Normal University through social network advertisements to participate in a study on alcohol use. They were occasional drinkers (mean 13.95 ± 17.82 times per year), and scored less than 8 on the Alcohol Use Disorder Identification Test (AUDIT), indicating no signs of underlying alcohol use disorders. 33 All participants were right‐handed, had normal or corrected vision and had no history of substance addiction or psychiatric disorders. The participants were randomly assigned to the placebo group (41 participants, 18 males) and the alcohol group (40 participants, 23 males), without being aware of which group they had been assigned to. Prior to their arrival at the laboratory for the study, participants were instructed to abstain from alcohol for 24 h and to refrain from coffee, cigarettes, water and food for 1.5 to 2 h. Each participant received 40 RMB as a reward. All participants were informed and consented to the study. The study was approved by the Ethics Committee of Hunan Normal University. MATERIALS The demographic information and drinking characteristics of the participants were evaluated. The drinking history was evaluated using the 10‐item AUDTI, which assesses a series of drinking habits. 33 Higher scores on the scale indicate a more severe alcohol use disorder. In addition, the items on the scale, namely ‘How often do you drink?’ and ‘How much do you generally drink?’, were used to ascertain the frequency and general quantity of drinking. In addition, participants were evaluated for depression, anxiety, and impulsivity. The Beck Depression Inventory, containing 21 items scored from 0 to 3, is employed to identify for depressive symptoms. 34 Higher scores on this scale indicate more severe depression. The Cronbach's α for this scale in this study was 0.823. The Beck Anxiety Inventory, containing 21 items scored from 1 to 4, is used to assess anxiety. 35 Higher scores on this scale indicate more severe anxiety. The Cronbach's α for this scale in this study was 0.820. The Barratt Impulsivity Scale, containing 30 items scored from 1 to 5, assesses unplanned impulsivity, motor impulsivity and attentional impulsivity in three dimensions. 36 Higher scores on the total and subscale scales indicate higher corresponding impulsivity. The Cronbach's α for the total scale in this study was 0.921. Alcohol administration Participants in the alcohol group were administered alcohol at a dose of 0.6 g/kg of body weight. This dose well represented moderate alcohol intoxication 37 and significantly impaired duration perception, 8 thus allowing for further exploration of the mechanisms of duration perception distortion. A total of 300 mL of mixed liquids was prepared, consisting of 46% white wine and lemon juice, and divided equally into two cups. The lemon juice was used to disguise the taste of the alcohol. Participants in the placebo group were similarly given 300 mL of mixed liquids that was divided equally into two cups, but which was almost entirely lemon juice. Only a little alcohol was sprayed onto the surface of the liquid and around the cup to induce an alcohol odour and prevent participants from guessing the experimental conditions. Participants were asked to drink all mixed liquids quickly (5 min maximum) in the presence of the researcher. Snacks were made available to participants during this period to alleviate the discomfort of acute alcohol use. Subsequently, participants gargled and were asked to engage in a total of 15 min of Sudoku puzzles and tasks familiarization to ensure that the alcohol exerted its physiological effects. 38 Throughout the study, researchers measured blood alcohol concentration (BAC) using the Mr·Black‐3 Breathalyzer, which automatically converts breath alcohol concentration to BAC. Procedure Participants were asked to complete relevant questionnaires for screening before arriving at the laboratory. After arriving at the laboratory, with informed and consent, they were first given a baseline BAC measurement to ensure that no alcohol was present in their system. They were then were administered either the placebo or alcohol, and waited for the alcohol to take effect. Whilst waiting, they engaged in Sudoku puzzles and practiced on the time bisection task, attentional network test and digit span backward test. After waiting, they first completed the subjective drinking experience questionnaires, then completed the formal phase of the corresponding task in turn, and finally completed the arousal assessment. BAC measurements were taken before each task and after completion of all tasks, totalling four times. To ensure the safety of the participants, they were not allowed to leave until their BAC was below 20 mg/100 mL. The subjective drinking experience questionnaires included the questions, ‘How much liquids containing alcohol do you feel you have drunk?’ (1 = not at all, 7 = very much) and ‘Do you feel inclined to get drunk?’ (1 = not drunk, 7 = very drunk). The arousal report contained the question ‘How aroused do you feel now compared to the calm state when you did not drink any liquids’ (1 = very slight, 9 = very strong). The time bisection task began with a learn phase, asking participants to memorize two standard durations of the short 300 ms and the long 1200 ms. 39 Each standard duration was presented five times at random, using the dark grey squares as carriers and with long or short cues. The practice phase then followed, where participants were asked to judge the length of the durations (300 and 1200 ms) without cues. After the same grey square presentation, participants needed to press ‘D’ in the keyboard if feeling that its duration was close to 300 ms. Conversely, they pressed ‘K’ if the duration was close to 1200 ms. Feedback was displayed after the response screen. The learn and practice phases were given to participants whilst they waited for the alcohol effects. So given the possibility of forgetfulness, a learn phase was again presented before the formal phase. The formal phase followed the same process as the practice phase, but with seven probe durations (300, 450, 600, 750, 900, 1050 and 1200 ms) to be judged and no feedback after the response. Each probe duration was presented 20 times for 140 trials in total, and they were randomly assigned to four blocks. The specific process is shown in Figure 1. The responses were recorded for each trial. Process of time bisection task. The attentional network test required participants to concentrate on quickly and accurately judging the direction of the target stimulus (the central arrow), after which they needed to press the appropriate key on the keyboard. 32 There are four cues: no cue, where only the fixation cross is displayed; central cue, where the fixation cross is replaced by an asterisk; dual cue, where two asterisks appear above and below the fixation cross; and spatial cue, where the asterisk appears above or below the fixation cross. None of the asterisks of the cues other than those of the spatial cue could predict the location of the target. In addition, flanks would appear on both sides of the target, and there were three types of directions of the flanks: consistent with the target direction (congruent condition), opposite to the target direction (incongruent condition) or two lines (neutral condition). Participants were asked to press ‘F’ when they observed the target directed to left and ‘J’ when the target was directed to right. The specific process is shown in Figure 2. The process was similar in the practice and formal phases, with the difference between feedback and no feedback. Four cues were combined with three flankers to form 24 trial types. These trials were repeated two times in the practice phase, for a total of 48 trials with feedback, and eight times in the formal phase, for a total of 288 trials without feedback, randomly assigned to four blocks. The RT and accuracy of each trial were recorded. Description of attentional network tests. (A) For the four possible cues; (B) for the six possible flankers; (C) for the trial example (neutral condition with double cue). Adapted from Fan et al. 32 The digit span backward test asked participants to hear a sequence of digits of increasing length over a series of trials and then to reproduce the digits in reverse order in the sequence order in which they appeared. 21 Participants were first presented with a sequence of digits from a list, and if they reproduced this sequence in reverse, they were presented with a longer sequence of digits. However, if they failed, they were presented with a sequence of digits of the same length again from an alternate list. If they succeed in reproducing it in reverse, then they continue to be presented with a longer sequence of digits from the original list. But if they fail in the alternate list as well, then the test is terminated. The number of digits in the longest digit sequence correctly reverse reproduced was recorded. Data analysis For the time bisection task, the proportion of ‘long’ responses was first calculated based on the ratio of responses judged to be ‘long’ to all responses at each probe duration. Furthermore, the point of subjective equivalence (PSE) was calculated based on the duration that accounted for 50% of the proportion of ‘long’ responses, which were used as the indicator of perceived duration. 39 The smaller PSE indicated the longer perceived duration. In addition, the Weber ratio (WR) was calculated based on the ratio of the standard deviation (SD) of each participants' PSE to their PSE, which was used as an indicator of sensitivity to time. 21 The smaller WR indicated greater sensitivity to time changes. For the attentional network test, data deletion was first performed. 32 Trials with incorrect responses and with RT below 200 ms or above 2000 ms were excluded. And trials in which each participant's RT was 2 SD below or above their individual average were also excluded. The subtraction method was then used to calculate the efficiency of the three attentional networks. 32 Alertness was calculated by subtracting the average RT of double cue trials from that of no cue trials, with higher values indicating higher efficiency. Orientation is the average RT for spatial cue trials minus that for central cue trials, with higher values indicating greater efficiency. Executive control is the average RT of incongruent condition trials minus that of congruent condition trials, with higher values indicating lower efficiency. For the digit span backward test, the more digits correctly reverse reproduced indicated better working memory capacity. 21 All statistical analyses were performed by SPSS 26.0. Demographic information was first analysed with independent samples t tests. A one‐sample t test was then performed on the scores of perceived alcohol in the placebo group. And a one‐way analysis of covariance (ANCOVA) was conducted on perceived drunk scores, with gender as a covariate. An ANCOVA was then performed on the BAC measured at the four time points (pre time bisection task, pre attention network test, pre digit span backward test and post all tasks) for the alcohol group, with gender as a covariate. The following analyses all incorporated gender, anxiety, depression and impulsivity as covariates. A 2 (alcohol manipulations: placebo and alcohol groups) × 7 (probe durations: 300, 450, 600, 750, 900, 1050 and 1200 ms) repeated measures ANCOVA was conducted on the proportions of ‘long’ responses. One‐way ANCOVA was used on the PSE and WR, respectively. In addition, a 2 (alcohol manipulation) × 3 (attentional networks: alertness, orientation, and executive control) repeated measures ANCOVA was conducted. And one‐way ANOVAs were conducted for working memory capacity and arousal, respectively. In addition, the mediating role of arousal and cognitive functions in the effect of acute alcohol use (as indicated by alcohol manipulation, setting the placebo group = 0 and the alcohol group = 1) on duration perception (as indicated by PSE) was examined, after controlling for gender, anxiety, depression and impulsivity. The mediating effects test was run using Model 4 of the Process plug‐in. Data included in this test were standardized. Bootstrap method was used to estimate 95% confidence intervals for the mediating effects by sampling 5000 Bootstrap samples. RESULTS The demographic information The placebo and alcohol groups did not differ in demographics statistics or drinking characteristics, as detailed in Table 1. TABLE 1Descriptive and inferential statistics for the placebo and alcohol groups (M ± SD).Placebo (n = 41)Alcohol (n = 40) t p Cohen's d Age20.85 ± 2.8120.88 ± 2.63−0.0350.9720.008Anxiety25.32 ± 4.1624.05 ± 3.861.4200.1590.320Depression4.46 ± 4.194.08 ± 4.170.4180.6770.094Impulsivity70.34 ± 16.5373.00 ± 13.91−0.7820.4360.176Alcohol use disorders3.68 ± 2.064.13 ± 1.84−1.0190.3110.229Drinking frequency1.51 ± 0.681.30 ± 0.461.6440.1040.370General alcohol use0.98 ± 1.041.17 ± 0.81−0.8500.3980.191Total drinking years4.27 ± 3.385.43 ± 2.85−1.6630.1000.374 Descriptive and inferential statistics for the placebo and alcohol groups (M ± SD). Subjective experiences of alcohol use The value of the placebo group on perceived alcohol (M = 2.59, SD = 0.95) was significantly different from 1 (representing no alcohol) (t (40) = 10.708, p < 0.001, Cohen's d = 3.386), meaning that the placebo group clearly sensed the presence of alcohol in the liquid consumed. In addition, the placebo and alcohol groups differed significantly in perceived intoxication (M = 1.59, SD = 1.07 vs. M = 3.83, SD = 1.69, F (2,78) = 50.557, p < 0.001, ηp2 = 0.393), implying greater perceived drunk in the alcohol group compared to the placebo group. Blood alcohol concentration All participants had zero BAC on arrival at the laboratory. The BAC in the placebo group was kept at zero. The alcohol group had similar BAC (F (3,37) = 1.005, p = 0.402, ηp2 = 0.075) prior to time bisection task (M = 48.25, SD = 14.31), prior to attentional network test (M = 49.83, SD = 15.84), prior to digit span backward test (M = 50.60, SD = 13.29) and after all tasks (M = 49.58, SD = 18.48). Duration perception On the proportions of ‘long’ responses (see Figure 3), the main effects of probe durations (F (6,70) = 46.513, p < 0.001, ηp2 = 0.799) and alcohol manipulations were significant (F (1,75) = 4.103, p = 0.046, ηp2 = 0.052), and the alcohol manipulation × probe duration interaction was marginally significant (F (6,70) = 2.082, p = 0.066, ηp2 = 0.151). The alcohol group had higher overall proportions of ‘long’ responses than the placebo group. Specifically, the alcohol group had higher proportions of ‘long’ responses compared to the placebo group at 450, 600 and 750 ms (ps < 0.05), whilst there were no differences between the two groups at 300, 900, 1050 and 1200 ms (ps > 0.05). Mean proportions of ‘long’ responses in the alcohol and placebo groups for each probe duration. Note. Vertical dashed and straight lines represented the points of subjective equivalence for the alcohol and comfort groups, respectively. On the PSE, a significant difference was observed between the placebo and alcohol groups (F (1,75) = 4.599, p = 0.035, ηp2 = 0.058; Figure 4A). The PSE in the alcohol group (M = 744.13 ± 85.26) was lower than that in the placebo group (M = 682.38 ± 110.84). However, no difference (F (1,75) = 0.090, p = 0.765) was observed between the placebo (M = 0.35 ± 0.12) and alcohol (M = 0.35 ± 0.15) groups on the WR. Mean values for the alcohol and placebo groups on (A) the point of subjective equivalence, (B) arousal, (C) working memory capacity, and (D) efficiency of attentional networks. Error bars are standard errors. ns, nonsignificant. *p < 0.05, *** p < 0.001. Arousal, attentional network and working memory capacity On the arousal, there was a significant difference between the placebo and alcohol groups (M Placebo = 5.17 ± 1.00 vs. M Alcohol = 8.30 ± 0.72, F (1,75) = 229.537, p < 0.001, ηp2 = 0.754; Figure 4B). This suggested that acute alcohol use increased arousal and accelerated pulse emission rate. On the attentional networks, the main effects of alcohol manipulation (F (1,75) = 2.183, p = 0.144) and attentional networks (F (2,74) = 1.624, p = 0.204) were not significant. However, the alcohol manipulation × attention network interaction was significant (F (2,74) = 10.155, p < 0.001, ηp2 = 0.215). Executive control was worse in the alcohol group compared to the placebo group (M Placebo = 68.24 ± 20.37 vs. M Alcohol = 85.82 ± 26.68, p < 0.001; Figure 4D), whilst there were no differences between the two groups on alertness (M Placebo = 37.73 ± 18.02 vs. M Alcohol = 30.06 ± 21.49, p = 0.110) and orientation (M Placebo = 39.93 ± 22.19 vs. M Alcohol = 43.05 ± 20.37, p = 0.839). On working memory capacity, there was a significant difference between the placebo and alcohol groups (M Placebo = 6.83 ± 1.55 vs. M Alcohol = 5.55 ± 1.69, F (1,75) = 14.368, p < 0.001, ηp2 = 0.161; Figure 4C), with the alcohol group having worse working memory capacity. Mediating effects test The mediating effects test showed a significant total effect (Effect = −0.461, SE = 0.215, 95% CI = [−0.890, −0.033]) and an insignificant direct effect (Effect = −0.075, SE = 0.415, 95% CI = [−0.754, 0.903]). In addition, the effects of alertness (Effect = −0.053, SE = 0.059, 95% CI = [−0.213, 0.019]), orientation (Effect = −0.009, SE = 0.049, 95% CI = [−0.099, 0.116]) and executive control (Effect = 0.080, SE = 0.088, 95% CI = [−0.098, 0.259]) were not significant, whilst the effects of arousal (Effect = −0.779, SE = 0.360, 95% CI = [−1.488, −0.065]; Figure 5) and working memory capacity (Effect = 0.225, SE = 0.119, 95% CI = [0.036, 0.496]; Figure 5) were significant. This suggested that arousal played a mediating role in the effect of acute alcohol use on duration perception, whilst working memory capacity served as a mask. Mediating model of the effects of acute alcohol use on duration perception. *p < 0.05, *** p < 0.001. DISCUSSION The present study examined the mechanism of the effect of acute alcohol use on duration perception from the perspective of cognitive functions. We found that acute alcohol use made individuals overestimate duration compared to placebo use, which supported our predictions and revealed the direction of the duration perception distortion. We also found that pulse emission rate mediated the effect of acute alcohol use on duration perception, whilst working memory capacity acted as a mask. The results of the PSE indicated that acute alcohol use could relatively result in overestimating duration, which is consistent with previous findings. 8 , 9 However, the results of the proportions of ‘long’ responses suggested that the effects of alcohol might vary for different probe durations. No significant differences were observed between the alcohol and placebo groups at 300, 1050 and 1200 ms, probably because these durations are closer to the standard short or long durations and therefore easier to judge accurately. 21 At 450, 600 and 750 ms, the alcohol group had higher proportions of ‘long’ responses than the placebo group, suggesting that they perceived these probe durations as closer to the long standard duration. Notably, at 900 ms, which approaches the middle probe duration (750 ms), the alcohol group had a similar response to the placebo group. These suggest that acute alcohol use may affect the perception of shorter durations (less than 750 ms) more than longer durations (more than 750 ms). However, the overall higher proportion of ‘long’ responses in the alcohol group compared to the placebo group still showed that acute alcohol use relatively resulted in overestimating duration. Arousal was found to be increased in the alcohol group. In addition, arousal mediated the effect of acute alcohol use on duration perception. Increased arousal implies that the pulse emission rate is accelerated. 3 Thus, accelerated pulse emission rate remained the cause of the relative overestimation of duration by acute alcohol use. 8 , 9 Alcohol accelerated the rate of pulse emission from individuals' pacemakers, allowing more pulses to pass through and accumulate in the accumulator, and thus, individuals showed the overestimation of duration. However, our findings also suggested that impaired working memory capacity, rather than any impaired attentional functions, contributed to the effect of acute alcohol use on duration perception. In terms of attention, acute alcohol use was found to impair executive control, but not alertness and orientation, which is consistent with previous research. 31 The alcohol inhibition model suggests that alcohol impairs the ability to inhibit rather than activate responses, 31 , 40 which may explain why alcohol only affected executive control. In addition, we found that the impaired executive control did not play any role in the effect of acute alcohol use on duration perception. This is similar to the findings in the relationship between alcohol use disorders and duration perception. 21 The role of attentional function in this mechanism is no longer evident due to the presence of other potential mechanisms (such as arousal and working memory). It seems more plausible that the executive control attention is not a necessary attentional function to perform the time bisection task (such as alert attention 17 ). In terms of working memory, acute alcohol use was found to result in poorer working memory performance, which is also consistent with previous findings. 30 In addition, we found that working memory capacity was also related to duration perception. The Cognitive Resource Allocation Model suggests that timing processes share limited memory resources with other cognitive processes. 19 Increasing the allocation of resources to the processing of other events could result in a loss of resources available for timing processes. 19 , 41 Consequently, due to the limited memory resources associated with timing, individuals were unable to maintain the amount of impulses that reflected the actual duration, thus shortening the perceived duration. Without considering whether resources have been allocated or not, it has been shown that poor working memory capacity itself affects the perception of duration. 20 , 42 Importantly, reduced working memory capacity was found to contribute to the underestimation of duration in patients with alcohol use disorder. 21 Thus, reduced working memory capacity itself could result in individuals underestimating duration. However, we found that working memory capacity played a masking rather than a mediating role in the influence mechanism. This might be attributed to an antagonistic effect between the working memory capacity and the pulse emission rate. Compared to placebo, alcohol intake itself leads to high arousal, an accelerated pulse emission rate and increased pulse storage, but simultaneously, working memory capacity is reduced and pulse storage is reduced. The partial reduced pulses did not allow the amount of pulses stored after acute alcohol use to be less than or similar to the amount of pulses stored after placebo use, so that the acute alcohol use ultimately still resulted in duration being overestimated, and the role played by the reduced working memory capacity was masked. The present study has contributed to clarifying the effect of acute alcohol use on duration perception and its mechanisms, but there are some notable limitations. Firstly, we used reported arousal to reflect pulse emission rate, but did not measure this indicator using the related task. Previous research had suggested that the slower finger tapping speed of depressed individuals reflected their reduced internal clock speed, thus explaining their underestimation of duration. 43 Additionally, it had been found that stimulant‐dependent individuals who overestimated duration exhibited increased finger tapping speed. 44 The finger tapping task could to some extent reflect the internal clock speed, 45 indicating the speed of the pulse emission rate. Future studies could consider measuring both duration perception and finger tapping speed after alcohol use to further validate our findings and interpretations. Secondly, previous studies employed different alcohol doses when exploring the acute effect of alcohol on duration perception, and the extend of cognitive functions impairment varied with alcohol dose or BAC. 29 , 30 Thus, the applicability of the antagonistic effects we found to other alcohol doses is unclear and could be further explored in the future. CONCLUSION In conclusion, our findings suggested that acute alcohol use relatively results in individuals' overestimation of duration. The primary factor contributing to the overestimation of duration is the increased arousal induced by acute alcohol use. However, the concomitant reduction in working memory capacity due to acute alcohol use masks the partial overestimation of duration. The present study provided insightful evidence to reveal the potential mechanisms underlying the effect of acute alcohol use on duration perception. CONFLICT OF INTEREST STATEMENT All authors declare that they have no conflicts of interest.