Website Loading Animation and Perceived Waiting Time: The Role of Temporal Attention

Abstract

The persistent challenge of designing digital interfaces that minimize users’ perceived waiting time remains critical for user satisfaction and conversion rates. This research integrates Attentional Gate Theory to investigate how loading animation type (static vs. dynamic) influences perceived waiting duration through temporal attention focus, and examines moderating roles of task involvement and browsing goal orientation. Across four online experiments—Study 1 (N = 198, MTurk) tested the main effect of animation type; Study 2 (N = 411, Prolific) validated full mediation via temporal attention focus using PROCESS analysis; Study 3 (2 × 2 design, N = 400, Prolific) examined task involvement as a moderator; and Study 4 (2 × 2 design, N = 400, Prolific) explored hedonic versus utilitarian browsing goals—the dynamic animation consistently shortened perceived waiting time relative to static displays. Mediation analyses confirmed that reduced temporal attention focus fully explains this effect, which is amplified under low task involvement and hedonic browsing but attenuated when involvement is high or goals are utilitarian. Theoretically, this work extends Attentional Gate Theory to user experience design by uncovering cognitive processes underlying time perception during waits. Managerially, it offers evidence-based recommendations for tailoring loading animations to user context, ultimately enhancing satisfaction and reducing abandonment.

1. Introduction

In the fast-paced new media environment, the consumer journey is defined by a series of micro-interactions, where even milliseconds can dictate engagement or abandonment. With user expectations for page load times narrowing to as little as two seconds [1], managing the subjective experience of waiting has become a frontier for interactive marketing innovation. Industry studies indicate that visual feedback during loading can substantially shape perceived waiting time. For instance, Harrison (2007) [2] found that dynamic animations, such as backward-flowing ribbing on progress bars, can make wait times appear 11% shorter. Similarly, interactive loading animations have been shown to reduce perceived wait time and enhance user satisfaction, particularly during longer delays [3]. However, visual content may have the opposite effect under very short wait times, increasing perceived delay [3], highlighting potential boundary conditions. More recent work demonstrates that active waiting tasks, varied visual indicators, and contextual factors such as salience, framing, and music influence both perceived duration and enjoyment [4,5,6]. These mixed findings indicate that prior results are not fully consistent, motivating further investigation. Collectively, these findings underscore that perceived waiting time is shaped not merely by elapsed seconds but by underlying psychological mechanisms, making it a pivotal area for research that bridges user experience (UX) design with consumer behavior.

Perceived waiting time, defined as users’ subjective assessment of how long they wait for a page to load, often diverges from objective system-recorded durations because it is influenced by attention allocation, emotional states, and contextual cues [7]. This construct is invaluable for interactive marketing research, as it reflects the psychological processes that mediate such effects. By contrast, other metrics—such as click-through behavior or delay tolerance—capture downstream interactions or stable trait-like limits rather than the dynamic, context-sensitive experience of waiting itself [3,8]. Therefore, it serves as a precise observational dimension for studying how marketing stimuli, such as loading animations, function in the new media landscape.

Among interface design features, the type of loading animation—static (e.g., stationary spinner) versus dynamic (e.g., moving or interactive progress indicator)—may subtly guide or retain user attention. This aligns with a core theory of time perception, particularly the Attentional Gate Theory [7], which posits that the more attention individuals allocate to the passage of time, the longer they perceive it to be. Applying this theory to a marketing context is an innovative step: it reframes the loading animation not as a technical feature but as a strategic marketing instrument designed to modulate consumer attention. Loading animations may thus act as a psychological lever, modulating temporal attention and thereby influencing perceived waiting time. Supporting this perspective, Lee et al. (2017) [9] found that distractor elements such as animated progress bars can shift attention and reduce perceived wait time, while Borges et al. (2015) [10] showed that emotionally engaging distractors, like in-store TV screens, shorten perceived waits and increase satisfaction. Empirical studies further indicate that features such as throbber speed, size, and duration [11], as well as motion speed, semantic cues, and graphic format [12], meaningfully affect perceived duration and user enjoyment. Similarly, consumer behavior research demonstrates that attentional allocation shapes evaluations and choices: attentional selection can increase product preference and perceived value [13], and focused attention on identity-related cues influences purchase decisions and consumption experience [14].

These convergent findings suggest that responses to waiting are not passive but mediated through cognitive attention mechanisms [15]. Loading animations, therefore, serve not merely as aesthetic flourishes but as functional stimuli that guide consumers’ attentional allocation. This insight aligns with recent consumer behavior research emphasizing the role of attentional focus in shaping subjective time estimates [16,17]. Specifically, temporal attention focus—the extent to which a consumer is cognitively attuned to the passage of time—has emerged as a key mediator explaining why different loading animations produce divergent experiences. For example, when participants engage with distractors such as familiar music or working memory tasks, they report shorter perceived waits compared to conditions in which attention is focused on time [18,19], implicating attentional diversion as a core mechanism.

Building on these insights, this study positions temporal attention focus as a central mediating variable linking loading animation type to perceived waiting time. While prior research has examined distractors broadly, few studies have explicitly tested the attentional mechanisms that explain their effectiveness in a new media marketing context. Even fewer have explored how specific animation types influence where and how consumers’ attention is allocated during waiting. Additionally, the moderating roles of factors like task involvement and browsing goals (hedonic vs. utilitarian)—cornerstones of marketing segmentation—remain underexplored, despite their theoretical significance. For instance, consumers with high task involvement may focus on functional outcomes and be less susceptible to attentional diversion, potentially weakening the effect of animation type on perceived waiting time. Similarly, browsing motivation may shape attentional priorities: hedonic browsers are likely more receptive to engaging visual stimuli, whereas utilitarian browsers prioritize efficiency and goal completion [20].

Despite growing interest, several gaps hinder the strategic use of loading animations in interactive marketing. First, few studies have systematically compared static versus dynamic loading animations within a controlled theoretical framework [9,10]. Second, the mediating role of temporal attention focus, though conceptually recognized, remains empirically under-tested. Third, crucial marketing variables like task involvement and browsing goal have not been integrated into models of wait perception. Addressing these gaps is essential for advancing both scholarly theory and the practical design of digital marketing strategies.

Building on the theoretical and empirical groundwork, this research explores three central questions:

RQ1: Does the type of loading animation (static vs. dynamic) influence consumers’ perceived waiting time?

RQ2: Is this relationship mediated by consumers’ temporal attention focus during the loading period?

RQ3: Do task involvement and browsing goal (hedonic vs. utilitarian) moderate the effect of animation type on temporal attention and, ultimately, perceived waiting time?

To examine these questions, we propose a moderated mediation model grounded in Attentional Gate Theory. Specifically, we posit that loading animation type influences perceived waiting time through its effect on temporal attention focus. This mediating process is further moderated by two variables. First, task involvement is hypothesized to weaken the mediation effect for highly involved users, who are less prone to attentional diversion. Second, browsing goal orientation is expected to strengthen the mediation for hedonic users, who are more receptive to engaging visual stimuli. The resulting model takes the form: Loading Animation Type → Temporal Attention Focus → Perceived Waiting Time, moderated by Task Involvement and Browsing Goal.

This research makes several key contributions to the field of interactive marketing innovation in new media. Theoretically, it makes an innovative leap by applying Attentional Gate Theory to a critical micro-interaction in the digital consumer journey. By empirically validating temporal attention focus as the key mediator, it builds a crucial bridge between psychological theory and interactive marketing, offering a nuanced cognitive explanation for how visual design elements in new media shape consumer perception. Second, it advances consumer behavior literature by integrating classic marketing moderators (task involvement, browsing motivation) into a model of time perception. This provides a more comprehensive and strategically relevant framework, moving beyond a one-size-fits-all approach to a context-aware understanding of consumer response.

Practically, our findings provide actionable strategies for interactive marketing innovation. They empower marketers and UX designers to strategically deploy animations not just as aesthetic elements, but as tools to manage consumer attention, reduce perceived friction, and enhance the overall brand experience. The identification of moderators offers clear guidance for segmenting and personalizing user experiences—using dynamic animations for low-involvement or hedonic contexts (e.g., browsing) and simpler ones for high-involvement or utilitarian tasks (e.g., checkout). This work provides an evidence-based blueprint for optimizing a small but powerful touchpoint in the digital consumer journey, ultimately improving satisfaction, retention, and conversion rates.

2. Literature Review

2.1. Attentional Gate Theory

Attentional Gate Theory (AGT), introduced by Zakay and Block (1995) [7], posits that individuals perceive time based on attentional resources allocated to temporal signals. When attention is focused on the passage of time, more ‘pulses’ pass through an internal gate to an accumulator, leading to longer perceived durations. The theory conceptualizes time perception as a function of cognitive processing, where an internal pacemaker emits pulses that are gated by attention before being accumulated. Thus, the more attention diverted to timing, the greater the number of pulses counted, resulting in an elongated subjective experience of time. In consumer behavior research, AGT has been widely used to interpret how environmental stimuli modulate perceived waiting time, especially in digital contexts. Recent studies have explored the role of interface design, interactivity, and user engagement as factors that either divert or concentrate attention, thereby altering perceived wait durations [21,22]. Dynamic elements such as animations are particularly relevant, as they can either distract from or enhance time awareness depending on their characteristics.

AGT provides a theoretical lens through which the effect of non-task-related stimuli—like loading animations—on temporal perception can be systematically interpreted. It suggests that the type of animation influences how much cognitive resource is allocated to monitoring time, thus affecting perceived waiting time. Specifically, dynamic animations may divert attention away from time tracking, reducing perceived wait, while static visuals may allow for more temporal monitoring. This insight helps researchers understand how design elements function as modulators of psychological time. When applying AGT to the relationship between loading animation type (independent variable) and perceived waiting time (dependent variable), the theory posits that dynamic animations act as attentional distractors, effectively narrowing the attentional gate. This reduces the number of temporal pulses that pass through, thus shortening perceived duration. In contrast, static animations keep attention on the passage of time, allowing more pulses to be accumulated and elongating perceived waiting time. Empirical findings support this claim—Karabay et al. (2021) [21] identified that attentional focus alters not only the awareness of time but also the accumulation process, and Olivers and Meeter (2008) [23] emphasized that attentional dynamics are susceptible to visual stimuli, which can either trigger or suppress the gating mechanism. The attentional blink literature further substantiates how transient attention shifts impact time perception [21]. Consequently, animation type, by altering attentional allocation, becomes a key predictor of subjective waiting experiences. This demonstrates how AGT not only explains but also predicts consumer responses to interface design in digital services.

2.2. Loading Animation Type (Static vs. Dynamic)

Loading animation type refers to the visual characteristics of the animation displayed during website loading, with static animations showing a fixed image or icon and dynamic animations featuring movement or changing elements [24]. The distinction between static and dynamic animations is significant as it may influence how users allocate attention and perceive the passage of time during wait periods.

Research on visual attention and time perception provides a foundation for understanding how loading animation type may impact perceived waiting time [25]. Studies have demonstrated that dynamic visual stimuli tend to capture and hold attention more effectively than static stimuli [26,27]. This enhanced attentional engagement with dynamic visuals may alter subjective time perception. For example, Kreitz et al. (2016) [28] found that exposure time, rather than motion speed, was the key factor influencing detection of unexpected visual objects. This suggests that dynamic animations which sustain attention over time may lead to different waiting time estimates compared to static animations.

The allocation of attentional resources also plays a crucial role in time perception during waiting periods. Heitz and Engle (2007) [29] showed that individuals with higher working memory capacity were able to more quickly constrain their attention in visual tasks. This indicates that the ability to focus and sustain attention on loading animations may vary across users, potentially leading to individual differences in perceived waiting time. Furthermore, Drew et al. (2013) [30] demonstrated that increased attentional demands can alter how visual information is processed and maintained. In the context of website loading, this implies that more engaging dynamic animations may place greater demands on attentional resources compared to static animations.

However, there are some gaps in the current understanding of how loading animation type influences perceived waiting time, particularly regarding how subtle visual features—such as motion blur or background clarity—shape attentional focus and subsequent perception [31]. While existing research provides insights into attentional processes and time perception, few studies have directly examined the relationship between static versus dynamic loading animations and subjective waiting time estimates in digital environments. Additionally, the potential moderating effects of individual differences in attentional control and working memory capacity on this relationship remain underexplored.

2.3. Temporal Attention Focus

Temporal attention focus refers to the degree of cognitive resources that individuals allocate to monitoring the passage of time during waiting periods [32,33]. This construct represents the conscious awareness and mental tracking of temporal cues, where higher temporal attention focus corresponds to greater cognitive vigilance toward time-related information [34]. The intensity of temporal attention focus fundamentally determines how individuals subjectively experience duration, positioning it as a critical mediator in time perception processes.

Building on this foundation, loading animation type significantly influences temporal attention focus through attentional capture mechanisms. Dynamic animations provide engaging visual stimuli that divert cognitive resources away from temporal monitoring, effectively reducing the mental allocation toward time passage [32]. Conversely, static loading displays fail to capture sufficient attention, allowing users to maintain heightened focus on temporal cues and duration tracking [35]. This differential attentional allocation establishes temporal attention focus as a key mechanism through which interface design elements influence subjective time estimation [36].

The impact of temporal attention focus on consumer behavior extends far beyond simple time estimation, fundamentally shaping user satisfaction and engagement in digital environments. When temporal attention focus remains high, consumers become acutely aware of waiting periods, leading to increased frustration, reduced satisfaction, and higher abandonment rates [37,38]. Research demonstrates that heightened temporal attention focus amplifies the negative psychological impact of delays, as users actively monitor each passing moment and become increasingly impatient [39]. This conscious time monitoring creates a psychological burden that transforms even brief delays into perceived obstacles to goal achievement [40]. Conversely, when temporal attention focus is reduced through effective attentional diversion, consumers experience shorter perceived waiting times and demonstrate greater tolerance for system delays [35]. This reduced temporal awareness allows users to remain more positively engaged with the interface, maintaining their task focus rather than becoming preoccupied with waiting duration. The strategic manipulation of temporal attention focus through design elements thus emerges as a powerful tool for enhancing user experience and optimizing conversion rates in digital consumer environments [37]. Collectively, these insights highlight temporal attention focus as a pivotal psychological mechanism that mediates the relationship between interface design choices and consumer behavioral outcomes.

3. Hypothesis Development

3.1. Loading Animation Type, Temporal Attention Focus and Perceived Waiting Time

Building upon the theoretical foundations of Attentional Gate Theory (AGT), dynamic loading animations during website waiting times have the potential to significantly reduce users’ perceived waiting time. This theory posits that subjective time perception is influenced by the amount of cognitive resources allocated to monitoring the passage of time [7]. Specifically, AGT suggests that when individuals direct more attention to temporal signals, they perceive time as passing more slowly, which elongates subjective duration [26,41]. Therefore, the less attention users allocate to the passage of time, the shorter they perceive the wait, which is a key tenet of AGT.

When users encounter loading animations on websites, the visual characteristics of these animations can dramatically alter how they experience the wait. In particular, dynamic loading animations—characterized by movement, progression, or changing visual elements—serve as engaging perceptual stimuli that capture users’ attention, effectively diverting it away from monitoring the passage of time itself [9,42]. The involvement of dynamic visual elements such as motion and continuous changes in the animation provides a focal point that draws attention away from the mere waiting experience, thereby reducing the cognitive resources dedicated to temporal monitoring [16,18].In contrast, static animations, with their fixed visual displays and minimal changes, fail to divert users’ attention away from the waiting process. As a result, users remain more attuned to the passage of time, which increases their awareness of waiting and elongates the perceived duration of the wait [9,10]. This is consistent with findings from prior research, which have shown that the absence of engaging stimuli leads to greater focus on time itself, thereby resulting in a longer perceived waiting time [3]. Accordingly, we propose:

H1. 

Dynamic loading animations lead to lower temporal attention focus compared to static animations.

Building on this premise, temporal attention focus serves as the critical mediating mechanism between loading animation type and perceived waiting time. When users encounter dynamic loading animations, their attentional resources are captured by the changing visual elements, effectively reducing temporal attention focus—the degree to which cognitive resources are directed toward monitoring the passage of time [43,44]. This redirection of attentional resources toward the animation itself leads to reduced accumulation of temporal “pulses” in cognitive time estimation, a process described in Attentional Gate Theory as essential for time perception [28,45]. With fewer attentional resources dedicated to temporal monitoring, fewer time markers are processed and stored in working memory, resulting in the subjective impression of a shorter waiting period [30,46]. Conversely, static loading animations fail to sufficiently engage users’ attention, allowing more attentional resources to be directed toward time passage monitoring. When temporal attention focus remains high, users actively track the passage of time, resulting in the accumulation of more temporal markers and consequently longer perceived waiting durations [29,47]. This mediation effect highlights the pivotal role of attentional processes in shaping time perception during digital waiting experiences, extending beyond direct animation effects. However, extant research further indicates the importance of examining how these cognitive mechanisms operate within the specific context of online waiting scenarios. Therefore, we propose:

H2. 

Temporal attention focus mediates the negative effect of dynamic loading animations on perceived waiting time.

In summary, this theoretical framework explains how loading animation type influences perceived waiting time through the mediating role of temporal attention focus. Based on the Attentional Gate Theory and supporting evidence, the above two hypotheses are proposed.

3.2. Moderating Role of Task Involvement

Building on well-established theories in consumer behavior, task involvement emerges as a significant moderator in the relationship between loading animation type and temporal attention focus. The allocation of attentional resources is directly influenced by task involvement, as demonstrated in multiple studies examining cognitive processing during online experiences [48,49]. When consumers exhibit low task involvement, they possess more available attentional resources that can be directed toward peripheral stimuli in their environment [50]. This attentional flexibility makes consumers more susceptible to distraction by dynamic loading animations, which draw attention away from time estimation processes. As Gendolla and Richter (2006) [51] established, low task involvement typically results in decreased concentration on the primary task, allowing environmental elements to capture attentional resources more readily. In contrast, high task involvement creates a state of focused attention where cognitive resources are primarily dedicated to completing the focal task [52,53]. This concentrated focus serves as a protective mechanism against potential distractions from dynamic elements in the interface. The reduction in available attentional resources for monitoring temporal cues is more pronounced when dynamic animations are present, as they compete for limited cognitive capacity [54]. Research by Meneghetti et al. (2011) [55] further suggests that under conditions of high involvement, individuals demonstrate enhanced ability to filter out extraneous stimuli, thereby minimizing the negative impact of dynamic animations on temporal attention focus. Together, these insights suggest a clear interaction between animation type and task involvement in determining temporal attention.

Building on this premise, we propose that the strength of the negative relationship between dynamic loading animations and temporal attention focus varies systematically according to the level of task involvement. When consumers engage with low involvement tasks, such as casual browsing, their relatively abundant attentional resources allow dynamic animations to significantly disrupt temporal attention processes [56,57]. Conversely, during high involvement tasks, such as completing important transactions, consumers maintain stronger attentional control, significantly reducing the disruptive effect of dynamic animations on temporal attention [58,59]. This interaction effect aligns with established attentional control theories suggesting that motivation type and temporal attention focus, such that the negative effect of dynamic animation (versus static animation) on temporal attention focus is stronger when task involvement is low than when task involvement is high. Thus, we suggested that:

H3. 

Task involvement moderates the relationship between animation type and temporal attention focus, such that the negative effect is stronger when task involvement is low.

3.3. Hedonic vs. Utilitarian Browsing Goal

The cognitive and attentional demands in online waiting contexts have been extensively explored, particularly in understanding how consumers perceive time during website loading periods. Existing research indicates that temporal attention focus, or the extent to which individuals actively monitor the passage of time, plays a crucial role in shaping perceived waiting time when influenced by dynamic (versus static) loading animations [60,61]. However, the relationship between temporal attention and perceived waiting time is likely contingent upon consumers’ browsing goals, which can be better understood through the dual-process model [62]. According to this model, consumers engage in either System 1 processing—fast, automatic, and emotionally driven—or System 2 processing, which is slower, deliberate, and logical [63]. This dual-process framework is especially relevant in distinguishing how different browsing goals—hedonic versus utilitarian—moderate how consumers allocate their attention during the waiting experience, influencing their perception of time.

Specifically, hedonic browsing, characterized by leisure-driven exploration and enjoyment, fosters experiential and pleasure-driven interactions with websites, leading to System 1 processing [64,65]. During this type of browsing, consumers tend to become deeply absorbed in website content, leading to time distortion, a phenomenon where their awareness of time passage naturally diminishes as they focus on enjoyment rather than the waiting experience itself [66,67]. This reduction in temporal awareness occurs because hedonic browsing directs attentional resources toward enjoyment, rather than task completion or efficiency metrics such as loading times [68]. System 1 processing, which is quick and automatic, dominates this experience, making time less salient. In contrast, utilitarian browsing, which is goal-directed and focused on efficiency, is more likely to engage System 2 processing, where consumers deliberately track time to achieve their tasks, making them more aware of delays such as prolonged loading times [69,70].Since utilitarian browsers tend to maintain a heightened awareness of the passage of time and evaluate websites based on their instrumental value in completing tasks, they are more sensitive to the effects of temporal attention on perceived waiting durations [67,71]. This aligns with System 2 processing, where individuals deliberately monitor time, making them more susceptible to perceiving longer waiting times when they actively focus on the progression of loading. Together, these insights suggest that the relationship between temporal attention and perceived waiting time will manifest differently based on consumers’ browsing goals and the cognitive processes they engage in.

Building on this premise, we propose that browsing goal type moderates the relationship between temporal attention focus and perceived waiting time. For utilitarian browsers, whose primary concern is efficient task completion, heightened temporal attention directly translates into longer perceived waiting times as they actively monitor loading progress [72,73]. However, hedonic browsers, who are primarily motivated by enjoyment and experiential aspects of website interaction, naturally experience diminished time awareness due to their immersive browsing state [74,75]. This immersion creates a buffer against the negative effects of heightened temporal attention, as their attentional resources are predominantly allocated to enjoyment rather than time monitoring [76]. Consequently, the positive effect of temporal attention on perceived waiting time is attenuated when consumers engage in hedonic browsing (See Figure 1). Formally, we hypothesize:

H4. 

Browsing goal type moderates the relationship between temporal attention and perceived waiting time, such that the positive effect is weaker for hedonic browsing goals.

4. Methodology

4.1. Overview of Study

Across four experiments we examine how the type of loading animation influences users’ perceptions of waiting time and the underlying cognitive processes and boundary conditions. Study 1 (N = 198, MTurk) tested the main-effect hypothesis that a continuously moving animation would shorten perceived wait times relative to a static symbol, finding that dynamic motion significantly reduced subjective duration even when objective exposure was held constant. Study 2 (N = 411, Prolific) replicated this effect in an e-commerce checkout context and, via PROCESS mediation, showed that dynamic animations diverted attention away from the passage of time—fully mediating the reduction in perceived waiting time. Study 3 (2 × 2; N = 400, Prolific) introduced task involvement as a moderator and revealed that the benefit of dynamic motion is pronounced under low involvement but attenuated when users are highly engaged in the task, consistent with limited attentional resources. Finally, Study 4 (2 × 2; N = 400, Prolific) tested browsing goal (hedonic vs. utilitarian) as a boundary condition, demonstrating that dynamic animations yield greater reductions in perceived waiting time when users adopt a hedonic browsing mindset compared to a utilitarian one. All participants were required to complete the experiment on a desktop or laptop computer. Mobile and tablet users were excluded to minimize variability in visual presentation. Additionally, participants were prompted to enable full-screen mode and were monitored for browser focus during the loading phase. Responses from participants who failed environment or attention checks (e.g., switching tabs or failing to maintain focus) were excluded from analysis.

In all studies, we employed a non-probability sampling strategy, using convenience samples from MTurk (Study 1) and Prolific (Studies 2–4). These platforms were chosen for their ability to quickly recruit diverse participants who meet specific demographic criteria, making them ideal for experimental research. The decision to use U.S. participants was based on practical considerations and the relevance of the U.S. as a representative market for digital consumer behavior, given its high internet penetration and diverse user base.

To enhance statistical reliability, a power analysis was performed using G*Power 3.1, verifying that the number of participants in each study was sufficient for detecting effects of moderate magnitude (Faul et al., 2009) [77]. The study aimed for a participant count of approximately 100 per experimental group, which would offer an 80% chance of identifying effects of medium magnitude (Cohen’s d = 0.40) at a significance threshold of α = 0.05 (two-tailed). The manipulation check results for each study are detailed in Appendix A.

4.2. Study 1: Main Effect of Loading Animation Type

4.2.1. Purpose of Study 1

Study 1 tested the main-effect hypothesis that the type of loading animation shown during a brief online waiting period shapes users’ perceived waiting time. Building on research on time perception and attentional diversion, we predicted that a dynamic (continuously moving) loading animation would make the wait feel shorter than a static (non-moving) loading symbol because animated stimuli capture attention and provide richer visual stimulation, thereby reducing the cognitive salience of elapsed time.

4.2.2. Stimuli and Pretest of Study 1

As no existing stimulus set exactly matched the present manipulation, so two information-symmetric loading screens were created. Both screens displayed the text “We are preparing your content” in the same blue, 14-point font against a white background and remained on screen for 15 s. In the dynamic condition, the text was accompanied by a circular ring that rotated clockwise at a steady pace of 120° per second. In the static condition, the identical ring appeared but was frozen at the 12 o’clock position; no other visual differences were present. Thus, movement was the sole manipulated feature (see details in Appendix B).

A pretest recruited 160 U.S. adults from Prolific (48.1% female, M_age = 34.27, SD = 9.11). After viewing one of the two loading screens (between subjects) for 15 s, participants answered the manipulation-check item “To what extent did the loading screen appear dynamic/moving?” (1 = not at all, 7 = very much). Participants exposed to the dynamic animation perceived it as markedly more dynamic (M = 6.12, SD = 0.88) than did those who saw the static version (M = 2.14, SD = 1.02), t(158) = 26.30, p < 0.001. These results confirmed that the stimuli effectively operationalized the intended contrast while holding all other screen properties constant.

4.2.3. Procedure of Study 1

First, two hundred U.S. participants were recruited from Amazon Mechanical Turk in exchange for a small monetary payment. After eliminating two respondents who failed an instructional-manipulation check, the final sample comprised 198 individuals (50.50% female; M_age = 35.86, SD = 10.44). Participants were randomly assigned to view either the dynamic (n = 99) or static (n = 99) loading animation in a between-subjects design described as an “interface evaluation task.” In the dynamic condition, participants saw a rotating icon, while in the static condition, the icon remained stationary. The purpose of this manipulation was to examine how the type of animation impacts participants’ temporal attention focus and perceived waiting time.

Next, upon entering the study, respondents provided demographic information and then read a brief cover story indicating that the researchers were testing a new video-hosting platform. They were told that, while the video loaded, they would see a temporary screen. The loading screen—dynamic or static—was then displayed for a fixed 15 s; navigation buttons were disabled to ensure equal exposure. Immediately afterward, participants completed a 60 s unrelated word-scramble puzzle that served as a distraction task to separate the manipulation from the dependent measures. It also ensured that participants would not immediately rate their perceived waiting time, preventing any immediate biases.

Then, participants rated their perceived waiting time. Perceived waiting time was assessed with three 7-point items adapted from Novak, Hoffman, and Yung (2000) [78]: “The wait felt very long” (reverse coded), “Time seemed to pass quickly during the loading screen” (reverse coded), and “I felt I had to wait a long time” (1 = strongly disagree, 7 = strongly agree). Items were coded so that higher values reflected longer perceived waiting; their internal consistency was satisfactory (α = 0.86). The manipulation check from the pretest (“How dynamic did the loading screen appear?” 1 = not at all, 7 = very much) was included after the dependent variable. Finally, participants answered attention-check questions and were debriefed.

4.2.4. Results of Study 1

Manipulation check. A one-way ANOVA confirmed the success of the animation-type manipulation. Participants rated the dynamic loading screen as far more dynamic (M = 6.05, SD = 0.82) than the static screen (M = 2.20, SD = 1.12), F(1, 196) = 302.46, p < 0.001, η² = 0.61, indicating a very large effect and validating the stimulus design.

Perceived waiting time. A second one-way ANOVA revealed a significant main effect of loading animation type on perceived waiting time, F(1, 196) = 24.38, p < 0.001, η² = 0.12. Participants who viewed the dynamic animation reported shorter waits (M = 3.12, SD = 1.02) than those who viewed the static animation (M = 4.02, SD = 1.09). The medium-sized effect supports the hypothesis that motion cues attenuate subjective duration even when objective waiting time is held constant. Furthermore, demographic variables such as age (F(3, 194) = 1.72, p = 0.16, η² = 0.026), gender (F(1, 196) = 0.34, p = 0.56, η² = 0.002), and education level (F(4, 193) = 1.24, p = 0.29, η² = 0.025) had no significant effects, indicating that these factors did not influence the perceived waiting time.

4.2.5. Discussion of Study1

Study 1 demonstrated that a simple interface design choice—a rotating versus stationary loading symbol—significantly influenced users’ temporal perceptions. Consistent with attentional models of time estimation, dynamic motion diverted cognitive resources away from clock-monitoring, yielding lower perceived waiting times.

4.3. Study 2: Mediating Role of Temporal Attention Focus

4.3.1. Purpose of Study 2

Study 2 investigated whether the type of loading animation shown during an online checkout (static vs. dynamic) causally influenced consumers’ perceived waiting time and, if so, whether this effect was mediated by temporal attention focus. Building on theories of attentional capture in human–computer interaction, we predicted that a dynamic, continuously moving animation would draw attentional resources away from the passage of time, thereby shortening perceived waiting time relative to a static, non-moving indicator.

4.3.2. Stimuli and Pretest of Study 2

Because no validated stimuli existed, we designed two information-symmetric loading screens that differed only in motion. In the static condition, participants saw a circular icon with eight equally spaced dots that remained motionless for fifteen seconds while a progress bar advanced in three discrete jumps (0%, 50%, 100%). In the dynamic condition, the identical icon rotated clockwise at a constant speed of one revolution per second, while the same progress bar advanced in the same three discrete jumps. Thus, visual content, duration, color palette, and wording (“Processing your payment…”) were held constant; only motion varied (see details in Appendix C).

A pretest on Prolific (N = 160; 52% female; M_age = 31.4, SD = 9.7) assessed perceived animation dynamism with a single seven-point item (“This loading screen looked dynamic,” 1 = strongly disagree, 7 = strongly agree). Participants rated the dynamic version (M = 6.12, SD = 0.99) as far more dynamic than the static version (M = 2.58, SD = 1.26), t(158) = 20.08, p < 0.001, confirming the effectiveness of the manipulation.

4.3.3. Procedure of Study 2

Four hundred and twenty US participants (50.5% female, M_age = 34.92, SD = 11.83) were recruited from Prolific for an “e-commerce interface study” in exchange for US $1.90. Participants were randomly assigned to view either the static or dynamic loading screen (between-subjects).

After demographic questions, participants were told they would test a prototype checkout page. They clicked “Pay,” whereupon the assigned loading screen appeared for fifteen seconds. Immediately afterwards they completed a 20 s filler task judging three unrelated abstract paintings to mask the study purpose.

Perceived waiting time (DV) was measured with three seven-point items adapted from Novak, Hoffman, and Yung (2000) [78]: “The wait felt very long,” “Time passed slowly,” and “I felt I waited a lot” (α = 0.87). To measure the mediation of temporal attention focus, we developed a three-item scale to assess participants’ attention to the passage of time during the loading period. The items were: “I was thinking about how much time was passing,” “My thoughts were focused on the length of the wait,” and “I kept track of the seconds ticking by” (α = 0.88). These items were designed to capture participants’ level of cognitive focus on the passage of time. A seven-point Likert scale was used for responses (1 = strongly disagree, 7 = strongly agree), with higher scores reflecting greater focus on time. This scale was pre-tested for reliability (α = 0.88). An alternative-explanation variable—perceived visual appeal of the screen—was measured with three items (“attractive,” “visually pleasing,” “professionally designed,” α = 0.91) adapted from prior interface research. The manipulation check asked, “How dynamic was the loading animation you just saw?” (1 = not at all, 7 = extremely). Finally, an instructed-response attention check (“Please select ‘2’ here”) appeared; nine participants failed and were removed, leaving N = 411 (df = 409) for analysis.

4.3.4. Results of Study 2

Manipulation check. Manipulation checks confirmed successful induction (see

Table 1. Summary of Studies.

Study	Sample & Design	Manipulation	Main Findings	Effect Size
Study 1	N = 198, between-subjects (dynamic vs. static)	Rotating ring vs. stationary ring (15 s loading screen)	Dynamic animation significantly reduced perceived waiting time compared to static.	η2 = 0.12 (medium); manipulation check η2 = 0.61 (very large)
Study 2	N = 411, between-subjects (dynamic vs. static)	Rotating vs. stationary icon with progress bar (15 s checkout loading)	Dynamic animation reduced perceived waiting time, fully mediated by reduced temporal attention focus; effect not due to visual appeal.	η2 = 0.08 (main effect); mediation indirect effect ab = −0.59, 95% CI [−0.78, −0.42]; manipulation check η2 = 0.51
Study 3	N = 400, 2 × 2 design (animation × involvement)	Dynamic vs. static loading screen × high vs. low task involvement	Dynamic animation reduced perceived waiting time strongly under low involvement, but effect was attenuated under high involvement.	η2 = 0.10 (animation); η2 = 0.07 (involvement); η2 = 0.04 (interaction); manipulation check η2 = 0.71
Study 4	N = 400, 2 × 2 design (animation × browsing goal)	Dynamic vs. static loading screen × hedonic vs. utilitarian goal	Dynamic animation reduced perceived waiting time, especially under hedonic browsing; effect smaller under utilitarian goal.	η2 = 0.12 (animation); η2 = 0.02 (goal); η2 = 0.05 (interaction); manipulation checks η2 = 0.18, 0.17

for details).

Main effect on perceived waiting time. A second one-way ANOVA revealed a significant effect of animation type, F(1, 409) = 35.88, p < 0.001, η² = 0.08. Waiting felt shorter in the dynamic condition (M = 3.52, SD = 1.14) than in the static condition (M = 4.67, SD = 1.23). We simultaneously analyzed demographic variables such as age, gender, and education level, but found no significant main effects or interactions (all p-values > 0.05).

Mediation analysis. Hayes PROCESS Model 4 (5000 bootstraps, 95% CI) tested temporal attention focus as mediator while controlling for visual appeal. Animation type significantly influenced temporal attention focus (a = −1.64, SE = 0.13, p < 0.001); greater attention focus lengthened perceived waiting time (b = 0.36, SE = 0.05, p < 0.001). The indirect effect (ab = −0.59, SE = 0.09, 95% CI [−0.78, −0.42]) was significant, whereas the direct effect of animation type on perceived waiting time, controlling for the mediator and alternative explanation, dropped to non-significance (c′ = −0.09, SE = 0.08, p = 0.26). Visual appeal did not predict waiting time (b = 0.04, SE = 0.04, p = 0.31), and the indirect path via appeal was not significant (95% CI [−0.03, 0.07]). Together, these results supported full mediation by temporal attention focus.

4.3.5. Discussion of Study 2

Study 2 demonstrated that a simple interface change—a continuously moving versus static loading icon—significantly altered consumers’ subjective waiting experience. The dynamic animation curtailed perceptions of delay, and this benefit was fully explained by reduced attention to the passage of time rather than by greater screen attractiveness.

4.4. Study 3: Moderating Role of Task Involvement

4.4.1. Purpose of Study 3

Study 3 was designed to test whether the beneficial impact of a dynamic versus static loading animation on consumers’ perceived waiting time depends on how involved they are with the focal task. According to Kahneman (1973) [79], attentional resources are limited, and when individuals are highly involved in a task, they allocate more cognitive resources to that task, reducing the attention available for distractions like animations. Building on attentional resource theories, we predicted that a dynamic animation, by providing visual movement, would shorten subjective waiting time more strongly when task involvement is low—because attentional resources are free to be captured—than when involvement is high, where attention is already allocated to the task itself.

4.4.2. Stimuli and Pretest of Study 3

Independent-variable stimuli consisted of two 15 s HTML loading screens created for the experiment. The static version displayed a navy-blue rectangle with the word “Loading…” centered in white, unmoving text. The dynamic version showed the identical rectangle and text, but the three dots cycled sequentially every 500 ms, accompanied by a subtle clockwise rotation of a 24 pixel ring. File size (220 kB), resolution (1080 × 720), color palette, and textual content were identical, ensuring information symmetry (see details in Appendix D).

Task involvement was manipulated through the written instructions preceding the loading screen. In the high-involvement condition participants read that they were beta-testing a soon-to-launch productivity app whose developers relied on their detailed evaluations; they were asked to scrutinize every feature carefully. The low-involvement instructions explained that the study merely calibrated screen brightness and that their responses were anonymous and non-consequential. Both texts were 110 ± 2 words, matched on Flesch–Kincaid readability (grade = 8.2) to eliminate confounds. We carefully ensured that the language used in both the task involvement and browsing goal primes was neutral and clear. The wording was designed to be simple and unambiguous, preventing any potential bias that might affect participants’ responses. This approach minimized the risk of wording effects, ensuring that the primes were interpreted as intended across participants.

A pretest with 160 U.S. adults recruited from Prolific (M_age = 40.2, 53% female) randomly assigned 80 respondents to the two animation versions and the other 80 to the involvement passages. Animation manipulation strength was measured with a single seven-point item (“This loading screen seemed animated and dynamic”); involvement with the three-item, seven-point scale from Avnet, Laufer, and Higgins (2013; α = 0.89). Dynamic screens were perceived as more dynamic (M = 6.12, SD = 0.82) than static screens (M = 2.04, SD = 1.01), t(158) = 25.90, p < 0.001. High-involvement instructions yielded higher involvement (M = 5.98, SD = 0.88) than low-involvement instructions (M = 2.31, SD = 1.13), t(158) = 23.87, p < 0.001. Thus, both manipulations operated as intended.

4.4.3. Procedure of Study 3

This study employed a 2 (animation: dynamic vs. static) × 2 (task involvement: high vs. low) between-subjects design. Four hundred U.S. adults were recruited from Prolific in March 2025 (52.5% female; M_age = 38.76, SD = 11.94) and randomly assigned to one of four cells (n = 100 each).

Next, participants provided informed consent before beginning the study. They were then presented with a task involvement manipulation. In the high task involvement condition, participants read a passage describing a beta-testing task for a productivity app, where they were told they would evaluate the app’s features carefully, providing detailed feedback. The instructions emphasized that their feedback would play a key role in improving the app’s usability. In contrast, participants in the low task involvement condition were told they were performing a routine calibration task, where their feedback was non-consequential and would not affect the app’s development. The difference in instructions was designed to manipulate the level of cognitive engagement with the task and test its moderating role on the effect of animation type. Then, all participants viewed a 21 s video, which ostensibly represented the app’s initial loading period. The video began at second 0 and ended at second 15, during which participants saw either a dynamic animation (a rotating ring) or a static animation (a fixed progress bar). The remaining six seconds were covered with a gray mask to standardize the total exposure time and prevent participants from focusing on anything beyond the loading animation. This ensured that the only difference across conditions was the type of animation shown, while the exposure time remained constant for all participants. After viewing the loading screen, participants were asked to complete a 30 s word-scramble task. This filler task was included to reduce carry-over attention effects from the animation and ensure that participants’ responses to the perceived waiting time were not influenced by the immediate visual stimuli.

Task involvement was measured with the Avnet et al. three-item scale (α = 0.91). The dependent variable, perceived waiting time, was captured with three seven-point items adapted from Novak, Hoffman, and Yung (2000) [78]: “The loading period felt very long,” “I was impatient while waiting,” and “Time seemed to drag during the wait” (α = 0.90). Similar procedures were used to measure temporal attention focus. Specifically, the same three-item scale was employed to assess participants’ temporal attention during the wait lile study 2. Finally, participants completed demographics and an attention-filter item; eight inattentive respondents were replaced to maintain the target sample size.

4.4.4. Results of Study 3

Manipulation checks. Manipulation checks confirmed successful induction (see Table 1 for details).

Perceived waiting time. A 2 × 2 ANOVA revealed main effects of animation, F(1, 398) = 42.18, p < 0.001, η² = 0.10, and involvement, F(1, 398) = 28.44, p < 0.001, η² = 0.07, qualified by the predicted interaction, F(1, 398) = 16.07, p < 0.001, η² = 0.04. Simple-effects analyses showed that under low involvement, dynamic animation significantly shortened perceived waiting time (M = 3.12, SD = 0.96) relative to the static version (M = 4.83, SD = 1.22), F(1, 398) = 56.14, p < 0.001. Under high involvement, the difference was attenuated (M_dynamic = 3.25, SD = 1.01; M_static = 3.68, SD = 1.07), F(1, 398) = 4.23, p = 0.041. Figure 2 visualizes the crossover pattern. We simultaneously analyzed demographic variables such as age, gender, and education level, but found no significant main effects or interactions (all p-values > 0.05).

Temporal Attention Focus. A 2 × 2 ANOVA on temporal attention focus revealed significant main effects of animation type (F(1, 398) = 28.45, p < 0.001, η² = 0.07) and task involvement (F(1, 398) = 24.51, p < 0.001, η² = 0.06), with a significant interaction (F(1, 398) = 10.34, p = 0.001, η² = 0.03). Under low involvement, dynamic animations (M = 4.56, SD = 1.21) captured significantly less temporal attention compared to static animations (M = 5.12, SD = 1.37; F(1, 398) = 21.76, p < 0.001). However, when task involvement was high, the difference was not significant (M_dynamic = 4.83, SD = 1.18; M_static = 4.90, SD = 1.20; F(1, 398) = 0.67, p = 0.414).

Moderated Mediation Analysis. A moderated mediation analysis using PROCESS Model 7 revealed a significant index of moderated mediation (b = 0.432, SE = 0.218, 95% CI [0.0125, 0.8762]). In the low task involvement condition, temporal attention focus significantly mediated the effect of animation type on perceived waiting time (indirect effect: b = −0.432, SE = 0.151, 95% CI [−0.7264, −0.1381]). In contrast, the indirect effect was not significant under high involvement (b = −0.091, SE = 0.146, 95% CI [−0.3767, 0.1945]; see Figure 3 for details).

4.4.5. Discussion of Study 3

Study 3 demonstrated that the effectiveness of dynamic loading animations in shortening perceived waiting time depends on consumers’ task involvement. When involvement was low, a moving animation substantially reduced subjective waiting duration, whereas the effect nearly vanished when involvement was high.

4.5. Study 4: Moderating Role of Browsing Goal

4.5.1. Purpose of Study 4

Study 4 was designed to investigate whether the effect of loading-animation type on consumers’ momentary browsing goal is contingent on task involvement. Building on dual-process perspectives of online information processing, we predicted that a dynamic (vs. static) loading animation would heighten a hedonic browsing goal, but primarily when consumers were weakly rather than strongly involved in the forthcoming task.

4.5.2. Stimuli and Pretest of Study 4

Independent-variable stimuli replicated those used in Study 2. Both versions lasted 15 s, occupied the same 1080 × 720 pixel frame, and displayed the text “Loading…”. In the static condition the text remained motionless; in the dynamic condition the three dots pulsed sequentially every 500 ms while a 24-pixel ring rotated slowly. File size, color, and wording were identical, ensuring information symmetry.

The moderator was operationalized with a browsing-goal prime adapted from Dhar and Wertenbroch (2000) [80]. Participants read an ostensibly unrelated instruction: “Imagine you are visiting an online store to (a) enjoy exploring fun products and be entertained” (hedonic) or “(b) quickly find a necessary product and complete the purchase efficiently” (utilitarian). They then wrote three sentences describing how they would browse under the assigned goal. The prompt length, word minimum, and time limit (90 s) were held constant across conditions.

A pretest with 160 U.S. adults from Prolific (M_age = 36.14, SD = 11.21; 53% female) confirmed both manipulations. Participants were randomly assigned to one of the two animations (n = 80 each) and rated perceived dynamism on a single seven-point item (“The loading screen felt dynamic”; 1 = strongly disagree, 7 = strongly agree). The dynamic version (M = 6.12, SD = 0.77) was judged more dynamic than the static version (M = 2.31, SD = 1.02), t(158) = 24.37, p < 0.001. A separate but equivalently sized sample evaluated the browsing-goal prime on a single item (“My current browsing goal is primarily for enjoyment”; reverse-scored for utilitarian). Hedonic priming produced higher enjoyment orientation (M = 5.89, SD = 0.83) than utilitarian priming (M = 2.54, SD = 1.04), t(158) = 21.46, p < 0.001.

4.5.3. Procedure of Study 4

The main study employed a 2 (loading-animation type: static vs. dynamic) × 2 (browsing goal: hedonic vs. utilitarian) between-subjects design. Four hundred U.S. panelists were recruited via Prolific (51.0% female, M_age = 35.72, SD = 12.08) and randomly allocated to the four experimental cells (n = 100 each). After providing informed consent and being told the research concerned “website usability,” participants completed the browsing-goal writing prime. A 15 s math-puzzle distraction task followed to reduce hypothesis guessing. Next, they viewed the assigned 15 s loading screen that ostensibly preceded product images.

Immediately afterward, they answered the manipulation checks: “The loading screen felt dynamic” (animation check) and “My browsing goal while viewing the site was mainly to have fun” (goal check), both on seven-point scales. Perceived waiting time, the dependent variable, was measured with three items adapted from Novak, Hoffman, and Yung (2000) [78]: “The wait felt short,” “Time passed quickly during the loading screen,” and “The delay seemed lengthy” (reverse-scored). Responses used seven-point agreement anchors; the scale was reliable (α = 0.86). Similar procedures were used to measure temporal attention focus. Specifically, the same three-item scale was employed to assess participants’ temporal attention during the wait lile study 2. Participants then completed demographic questions and an attention-filter item.

4.5.4. Results of Study 4

Manipulation checks. Manipulation checks confirmed successful induction (see Table 1 for details).

Perceived waiting time. A 2 × 2 ANOVA on the composite waiting-time index revealed a significant main effect of animation type: dynamic screens were perceived as shorter (M = 2.68, SD = 1.02) than static screens (M = 3.22, SD = 1.04), F(1, 396) = 52.61, p < 0.001, η² = 0.12. The main effect of browsing goal was also significant, F(1, 396) = 8.53, p = 0.004, η² = 0.02, with hedonic browsers reporting slightly shorter waits (M = 2.75) than utilitarian browsers (M = 3.15). Crucially, the Animation × Goal interaction emerged, F(1, 396) = 19.84, p < 0.001, η² = 0.05. Simple-effects analyses showed that under a hedonic goal, dynamic animations (M = 2.30, SD = 0.90) led to markedly shorter perceived waiting times than static animations (M = 3.20, SD = 1.05), F(1, 396) = 55.77, p < 0.001. Under a utilitarian goal, the reduction was smaller (M_dynamic = 3.05 vs. M_static = 3.25), F(1, 396) = 4.11, p = 0.043 (see Figure 4 for details). We simultaneously analyzed demographic variables such as age, gender, and education level, but found no significant main effects or interactions (all p-values > 0.05).

Temporal Attention Focus. A 2 × 2 ANOVA on temporal attention focus revealed significant main effects of animation type (F(1, 396) = 30.14, p < 0.001, η² = 0.07) and browsing goal (F(1, 396) = 6.92, p = 0.009, η² = 0.02), with a significant interaction between the two factors (F(1, 396) = 15.13, p < 0.001, η² = 0.04). Under a hedonic goal, dynamic animations (M = 4.41, SD = 1.23) captured significantly less temporal attention than static animations (M = 5.15, SD = 1.22; F(1, 396) = 21.22, p < 0.001). However, under a utilitarian goal, the difference was less pronounced (M_dynamic = 4.92, SD = 1.17; M_static = 5.05, SD = 1.15; F(1, 396) = 1.12, p = 0.291).

Moderated Mediation Analysis. A moderated mediation analysis using PROCESS Model 7 revealed a significant index of moderated mediation (b = 0.527, SE = 0.256, 95% CI [0.0374, 1.1034]). In the hedonic browsing goal condition, temporal attention focus significantly mediated the effect of animation type on perceived waiting time (indirect effect: b = −0.524, SE = 0.198, 95% CI [−0.9143, −0.1512]). In contrast, the indirect effect in the utilitarian condition was not significant (b = −0.115, SE = 0.142, 95% CI [−0.3937, 0.1654]).

4.5.5. Discussion of Study 4

Study 4 demonstrated that the beneficial effect of dynamic loading animations on perceived waiting time is not universal but depends on users’ browsing goals. When participants browsed for enjoyment, the moving dots and rotating ring significantly accelerated subjective time, whereas utilitarian browsers derived only modest benefit.

5. General Discussion

This research investigated the impact of loading animation type (static vs. dynamic) on perceived waiting time in online environments across four studies. The findings consistently show that dynamic loading animations can significantly reduce users’ perceived waiting time compared to static animations, with important boundary conditions and underlying mechanisms identified [3,7]. Study 1 established the main effect, showing that participants who viewed a dynamic loading animation reported shorter perceived waiting times than those who viewed a static animation, despite equal objective durations, addressing whether animation type influences perceived waiting time. This effect aligns with attentional models of time estimation, suggesting that dynamic motion diverts cognitive resources away from temporal monitoring [45]. Study 2 replicated this main effect and identified temporal attention focus as the mediating mechanism, with the dynamic animation reducing focus on the passage of time, which in turn shortened perceived waiting duration, demonstrating the mediating role of temporal attention focus [43,45]. Importantly, this mediation held even when controlling for the alternative explanation of visual appeal, indicating that the effect is driven by attentional processes rather than aesthetic preferences. Study 3 revealed task involvement as a crucial moderator, with the beneficial effect of dynamic animations being substantially stronger under low involvement but nearly absent under high involvement, highlighting the moderating role of task involvement [48,69]. This suggests that the effectiveness of dynamic animations depends on the availability of attentional resources that can be captured by the animation. Study 4 extended these insights by showing that users’ browsing goals also moderate the effect of animation type, with dynamic animations being particularly effective for hedonic browsing goals while less pronounced for utilitarian goals, illustrating the moderating influence of browsing goals [10].

5.1. Theoretical Contribution

This research extends the application of Attentional Gate Theory in the field of digital consumer behavior [61], providing a new theoretical perspective on how interface design can influence users’ time perception by impacting cognitive processes, particularly attention allocation. By investigating how loading animation type (static vs. dynamic) influences perceived waiting time through temporal attention focus, this study fills a theoretical gap between interface design and time perception mechanisms. First, anchored in Attentional Gate Theory [7], our study extends prior research on attentional diversion by specifying how different interface visual designs—particularly animation types—modulate temporal attention focus and perceived waiting time. By empirically validating temporal attention focus as a mediator [41,45], we bridge the gap between user interface design principles and psychological theories of time perception. While previous studies have examined attentional diversion in various contexts [9,10], our research goes further by integrating these insights within a digital interface context and testing the boundary conditions under which such attentional diversion is most effective. Our findings demonstrate that dynamic loading animations can reduce perceived wait time through attentional diversion [3], supporting the strategic use of animation design to enhance user experience.

Second, we advance consumer behavior literature by integrating individual-level (task involvement) and situational (browsing motivation) moderators [48,64], thereby offering a more comprehensive model of perceived waiting time in digital environments. By integrating factors such as task involvement and browsing goal, we provide a more comprehensive model that explains how the effects of loading animations vary according to user engagement and motivational type. The study shows that, under low task involvement and hedonic browsing contexts, dynamic animations are more effective, while the effect is weakened under high task involvement or utilitarian browsing. This theoretical expansion offers new perspectives for future research on interface design and user experience, particularly regarding the time perception differences between task-oriented and entertainment-oriented users.

Third, we extend previous work on distractors and time perception in digital interfaces by operationalizing animation type and attention mechanisms in a unified framework [21,81]. We show that attentional gates can be strategically modulated through interface design, thereby extending Attentional Gate Theory and clarifying how interface design shapes attentional allocation and the cognitive mechanisms underlying perceived waiting time. Our research provides a more granular understanding of how specific design elements (static vs. dynamic animations) influence cognitive processes (temporal attention focus) and ultimately shape user perceptions of waiting time. This contribution offers a theoretical foundation for future studies exploring the interplay between interface design, cognitive processes, and user experience in digital environments [82].

5.2. Practical Contribution

The findings of this study offer valuable, evidence-based guidance for web designers, UX professionals, and digital marketers aiming to optimize online waiting experiences. Dynamic loading animations consistently reduce perceived waiting time compared to static indicators, making them an essential tool in improving user satisfaction and reducing abandonment rates across digital platforms. On desktop platforms, designers should incorporate rotating or continuously moving elements, such as circular rings with progress indicators, to maintain attention during brief delays in scenarios like e-commerce checkouts, content streaming buffering, or dashboard loading screens. In mobile environments, waiting can be more disruptive due to smaller screens, variable bandwidth, and touch-based interactions. Subtle motion cues, short interactive elements, or simplified animations under low-bandwidth conditions can help maintain users’ attention without overwhelming device resources. However, designers should also consider potential trade-offs, such as increased load times or user distraction, and balance animation complexity accordingly. To further enhance user experience, designers can combine dynamic animations with complementary distractions, such as relevant content snippets, brand messaging, or progress indicators, to keep users engaged while waiting. By addressing specific constraints in both desktop and mobile environments, designers can maximize the effectiveness of dynamic animations, reduce perceived wait times, and improve user satisfaction. This will ultimately lead to reduced abandonment rates and higher user retention across diverse platforms.

Our research reveals that temporal attention focus serves as the critical mediating mechanism between loading animation type and perceived waiting time, offering valuable implementation guidance. User experience professionals can enhance waiting experiences by deliberately designing interfaces that redirect users’ attention away from time monitoring [83]. For example, dynamic animations should be visually engaging enough to capture attention but not so complex that they consume excessive processing resources, which could paradoxically increase loading times. Practitioners might consider implementing subtle animations that change direction or incorporate small progressive elements that provide a sense of advancement without requiring conscious attention. This approach becomes particularly effective when combined with complementary distraction techniques, such as displaying relevant content snippets or brand messaging alongside the animation. By understanding that temporal attention focus—not merely visual appeal—drives the effectiveness of dynamic animations, designers can create loading experiences that actively divert users’ cognitive resources away from time perception, resulting in more positive waiting experiences.

Our findings demonstrate that the effectiveness of dynamic loading animations varies significantly depending on users’ task involvement, informing optimal deployment strategies. The research highlights that dynamic animations are most effective when users exhibit low task involvement but produce diminished benefits in high-involvement scenarios. Digital marketers should therefore adapt loading animation complexity based on the user’s likely involvement level within the customer journey. For casual browsing contexts, such as product discovery pages or content exploration, elaborate dynamic animations can substantially reduce perceived waiting durations. Conversely, for high-stakes interactions like payment processing or form submissions, where user involvement is naturally elevated, designers should implement more subtle, professional animations that acknowledge the user’s focused attention. This differentiated approach contrasts with the common practice of applying uniform loading animations across all website sections. This differentiated approach helps avoid overloading users or creating distraction in critical tasks. In designing adaptive interfaces, practitioners should consider automating animation selection based on behavioral indicators of involvement—such as session duration, click depth, or interaction patterns—ensuring that animations are tailored to the context of use and maximize their effectiveness in enhancing user experience.

Our research has significant cross-contextual applications that transcend specific digital platforms and cultural contexts. The effectiveness of dynamic loading animations in modulating perceived waiting time through attentional diversion represents a universal psychological mechanism that operates across diverse user populations. This finding is particularly valuable for global digital services that must accommodate varying internet speeds and cultural expectations regarding wait times. Nonetheless, designers should weigh performance and cultural preferences when implementing animations to avoid unintended negative effects, such as frustration or distraction. For example, in regions with slower internet infrastructure, implementing culturally appropriate dynamic animations can mitigate user frustration during longer loading periods. Similarly, the interaction between browsing goals and animation effectiveness is universal—hedonic browsers consistently benefit more from dynamic animations than utilitarian users, regardless of cultural background. However, cultural calibration is still essential. What is considered an appropriately engaging dynamic animation may vary between regions. For instance, Western users may prefer minimalist rotating indicators, while users in certain Asian markets may respond better to character-based animations that convey personality. By understanding these nuances, global digital experience designers can implement culturally responsive, yet psychologically grounded, loading experiences that resonate with diverse markets.

5.3. Limitations and Future Research Directions

Despite its robust experimental design, this research presents several notable limitations. First, all studies relied on brief, artificial loading scenarios with fixed durations, raising questions about generalizability to more variable or naturally occurring online waits, where duration and context may alter user attention and perceptions [9]. Second, participant samples were exclusively drawn from U.S.-based online panels, which limits the cultural diversity of the data and raises concerns about generalizability, as time perception and attentional responses to interface design can vary substantially across cultural contexts [16]. Third, the operationalization of dynamic versus static animations was highly controlled, focusing on minimal movement differences; real-world interfaces often present richer, more complex visual cues that could interact with additional user and situational factors [10]. Future research should replicate these findings in field settings with more diverse populations, explicitly including cross-cultural comparisons and mobile versus desktop contexts, and explore longer and more varied waiting experiences to enhance ecological validity. Future studies could also examine more complex or interactive animations and other interface elements, such as color or layout, to better understand how multiple design features jointly influence perceived waiting time. Collectively, such extensions would refine theoretical understanding of attentional mechanisms in digital waiting experiences and provide more actionable guidance for interface design [84].