Abstract
Syntactic priming is a linguistic phenomenon where the exposure to a specific grammatical structure increases the likelihood that the same structure will be used in subsequent language production. Previous research has shown that these syntactic processes are influenced by a broad range of linguistic, cognitive, and social factors. In the present research, we focused on the role of religious systems in shaping syntactic priming in dialog. Based on the observation that the Taoist concept of “moving with flow” encourages individuals to adapt to the natural flow of life and circumstances, we hypothesized that Taoists might modify their syntactic choices to foster better connections with their interlocutors. We conducted two studies to test our theoretical perspective. As predicted, Study 1 found that Taoists were more likely to mirror the syntax used by their interlocutors in dialog than their atheist counterparts. In Study 2, we compared the magnitude of syntactic priming effects between Taoists and Buddhists. The results showed that Taoists demonstrated a stronger preference for using the same syntactic structure as their conversational partners than Buddhists. In sum, our findings provide the first empirical evidence that beliefs and values associated with religious systems exert a significant influence on syntactic priming effects.
Adoption of others’ behaviors and social mimicry are fundamental to human evolution because they facilitate learning, social bonding, and the transmission of culture (Chartrand & Van Baaren, 2009; Stel & Vonk, 2010). By mimicking others, individuals acquire skills and knowledge that are crucial for survival and social interaction (Lakin et al., 2008). One paradigmatic example of social mimicry in linguistic behavior is syntactic priming, in which exposure to a particular sentence structure influences the construction of subsequent sentences during conversation (Branigan et al., 1995; Pickering & Branigan, 1999; Tooley & Traxler, 2010). For instance, frequent exposure to the sentence “She enjoys reading books” might increase the likelihood of adopting a similar syntactic structure, such as “He loves playing soccer,” in subsequent discourse.
Linguists and cognitive scientists have proposed several theoretical accounts to elucidate the underlying mechanisms of syntactic priming. According to one such theory, the residual activation account suggests that when a specific syntactic structure is encountered, the associated combinatorial nodes remain “activated” for a period even after the structure has been processed (Dell, 1986; Pickering & Branigan, 1998). This residual activation may enhance the likelihood of reusing the same syntactic structure over alternative ones. For instance, following exposure to an active sentence such as “The chef cooked a delicious meal,” a speaker may be inclined to formulate a new sentence that maintains the same voice. In this instance, they might construct the sentence “The artist painted a beautiful picture,” instead of choosing a passive structure like “A beautiful picture was painted by the artist.”
A critical assumption of the residual activation account posits that the syntactic influence is transient, not lasting across numerous sentences. However, accumulating evidence suggests that exposure to a specific syntactic structure can exert a prolonged influence on subsequent language production over extended dialogs (Bernolet et al., 2016; Thothathiri & Levshina, 2023). For instance, Bock and Griffin (2000) demonstrated that syntactic priming effects could endure beyond immediately adjacent sentences, persisting over 10 intervening sentences. These long-lasting effects have given rise to an alternative theoretical framework, the implicit learning accounts, which attribute priming to an error-based implicit learning mechanism.
The implicit learning account posits that syntactic priming relies on the observation and reinforcement of syntactic structures during language use, especially within contexts where individuals unconsciously adjust to communicative errors (Chang et al., 2006; Jaeger & Snider, 2013). Specifically, speakers make predictions based on past linguistic encounters, adjust their assumptions upon confronting discrepancies, and incrementally refine their linguistic comprehension through sustained exposure and adaptability. This dynamic interplay between prediction, error detection, and subsequent adaptation fosters continuous learning and enhances syntactic priming (Jackson & Ruf, 2018; Kaan & Chun, 2018).
For example, consider an English learner who frequently encounters double object constructions (e.g., “She sent him a letter”). If they primarily anticipate hearing sentences structured in this manner and then come across a prepositional object construction (e.g., “She sent a letter to him”), their subsequent predictions may shift to accommodate this less frequent, unexpected structure. This mismatch encourages the learner to adjust their expectations, incorporating the structures they have encountered less often, thus cultivating a more nuanced and accurate syntactic model of English. Thus, primed production occurs not just from immediate activation but as a reflection of an evolving understanding of language patterns shaped by implicit learning from past experiences.
Regardless of whether the underlying mechanism is driven by activated representations (as posited by the activation account) or by mappings of meaning to structure (as suggested by the implicit learning account), both theoretical frameworks emphasize that the impact of previous exposures operates autonomously from the speaker's conscious control and external situational variables. Yet, an emerging body of research challenges this notion, which indicates that structural priming is not entirely independent but is modulated by sociocognitive factors. For instance, Balcetis and Dale (2005) observed enhanced syntactic priming effects when participants interacted with a favored conversational partner compared to when they interacted with a less favored one. More recently, Hwang and Chun (2018) corroborated these findings in a real-world interactional setting. They found that students were more likely to repeat a sentence structure of a teacher whom they viewed positively. Overall, these findings suggest that sociocognitive variables such as social perception significantly constrain syntactic use and mimicry.
Moving beyond mere correlational evidence, recent studies have explored the causal effects of social factors on syntactic priming, considering variables such as accent (standard vs. nonstandard; Chun & Kaan, 2022), the identity of speakers (high prestige vs. low prestige; Lev-Ari, 2016), and the nature of conversational agents (AI vs. Human; Cowan et al., 2015). For instance, reasoning that social ostracism intensifies the need to regain social acceptance and ameliorate feelings of isolation, Li (2023) examined whether exclusion could heighten language mimicry. As hypothesized, the findings indicated that individuals who experienced social exclusion exhibited a more pronounced propensity to emulate the syntactic structures employed by their interlocutors compared to their included counterparts. These findings converge to highlight the important role of nonlinguistic factors in influencing how interlocutors align linguistically with their conversational partners in communicative contexts.
Multiple sociolinguistic theories have been proposed to account for the influence of social factors on syntactic priming. One of the most influential theoretical frameworks is the Communication Accommodation Theory (CAT; Giles et al., 1973). According to this account, speakers may align their speech patterns, vocabulary, and accents with those of their conversation partners to foster affiliation. Alternatively, they may diverge their linguistic practices to denote social distance in contexts of social hierarchy, conflict, or personal disaffiliation (Giles et al., 1991, 2023). For example, Chun et al. (2016) found that American English-speaking participants exhibited reduced syntactic priming effects when exposed to American-accented speech compared to Korean-accented or Indian-accented speech. It may be that nonstandard accents prompt American-English speaking participants to accommodate their language more, thereby intensifying syntactic priming as they attempt to bridge the gap and connect with the speaker. Overall, this finding aligns with CAT, which posits that speakers and listeners may adjust their linguistic behavior based on social dynamics, including accent familiarity and perceived social distance.
In the current research, we explored the novel possibility that religious systems, or religions, may influence syntactic priming effects. Specifically, we examine the impact of the Taoist belief in “moving with the flow” on speakers’ syntactic choices in language production. This metaphor, which describes life's challenges as a river that flows around obstacles rather than confronting them, encapsulates the Taoist way of dealing with life's difficulties (Koenig & Spano, 1998). Inspired by this belief, Taoism promotes adapting to the natural rhythms of life, an idea that mirrors the concept of accommodation in communication (Wong, 1992). In particular, CAT describes adjusting language to strengthen social bonds and minimize social distance, which parallels Taoism's encouragement of living in harmony with others and the world.
In recent years, an increasing number of empirical studies have shown that this Taoist belief significantly affects cognitive processing (Hagen, 2002). Previous research has illustrated that religiosity and engagement in faith practices enhance task persistence, as the core doctrines of many religions (e.g., Christianity, Buddhism, Islam) advocate sustained effort over time despite adversity (Toburen & Meier, 2010). In contrast, Li (2024) suggests that Chinese Taoists, influenced by beliefs emphasizing nonvolitional action and wise passivity (“moving with the flow”), may exhibit less task persistence than followers of other religions. Supporting this, the researcher found that Taoist participants exposed to God-related primes demonstrated lower levels of task persistence compared to their Buddhist counterparts. Thus, these findings suggest that the “move with the flow” concept in Taoism significantly influences its adherents’ cognition and behavior.
More relevant to the present study, recent investigations have shown that this Taoist principle also affects language processing (Li & Cao, 2018). For instance, Li and Cao (2021) explored how Taoist philosophy affects its believers’ interpretations of temporally ambiguous statements. The researchers found that Taoists who emphasized passivity, calmness, and nonstrife were more likely to adopt a time-moving perspective when interpreting the ambiguous sentence (i.e., “If Wednesday's meeting is moved forward 2 days, is it now on Monday or Friday?”). This means that they are more inclined to respond with “Monday,” perceiving themselves as stationary while temporal events move toward them. Overall, these findings suggest that religious beliefs may be associated with distinct patterns of language and cognitive processing.
Despite initial evidence supporting the influence of Taoism on language processing, previous studies have predominantly concentrated on semantic aspects, such as temporal reasoning. It remains unclear whether the Taoist belief in “moving with the flow” affects syntactic choices in language production. As previously discussed, the concept of “moving with flow” in Taoist philosophy underscores the importance of aligning one's actions and responses with the natural rhythms and dynamics of the environment. In communicative contexts, this principle can be interpreted as a form of language accommodation, wherein individuals adjust their linguistic choices to harmonize with those of their conversational partners. Thus, this philosophical orientation may enhance sensitivity to contextual dynamics.
In terms of syntactic alignment, “moving with flow” suggests that speakers tend to subconsciously modify their syntactic structures to mirror those of their interlocutors. Such alignment is not merely mechanical imitation; rather, it reflects an intuitive responsiveness to the communicative context and the flow of conversation. For example, if one participant employs a complex sentence structure, a listener who embodies this Taoist philosophy may naturally adopt a similar structure in their subsequent responses.
Research has demonstrated that such syntactic priming effects can enhance mutual understanding and facilitate smoother interactions (Hopkins & Branigan, 2020; Li, 2023). By accommodating their language to align with the syntactic choices of others, speakers can establish a shared communicative ground, thereby promoting cohesion and reducing misunderstandings. Given Taoism's emphasis on environmental awareness and the fluidity of life, which parallels the adaptive communication strategies posited by CAT, we hypothesize that Taoism will correlate with a stronger syntactic priming effect. To test our theoretical perspective, we conducted two studies with high statistical power. In Study 1, we compared the magnitude of syntactic priming effects between Chinese Taoists and atheists. To eliminate the possibility that general religiosity accounts for the observed effects, Study 2 replicated the findings with two comparable groups of Taoists and Buddhists.
Study 1
Method
Participants
The study was conducted in Chengdu, the capital of Sichuan Province, located in southwestern China. Employing G*Power for statistical analysis, we determined that 36 participants were necessary to detect a medium effect size (f = 0.25) in a two-way repeated measures ANOVA with a power of 0.95. Anticipating potential participant attrition, we recruited 40 Taoists and 37 atheists given that time and research funding allowed. All Taoists had been practitioners for at least 6 years at the time of the study. They were sourced from local Taoist temples, primarily engaged in rituals aimed at promoting the welfare of the community and its residents. All atheist participants were recruited from the communities in the same city. They self-identified as atheists due to their absence of engagement in any faith-based practices.
Throughout the recruitment process, we meticulously matched participants in terms of various sociodemographic variables such as age, gender, geographic locations, and educational attainment (refer to Table 1). The Taoists declined remuneration, citing that acceptance would contravene specific ethical or doctrinal standards. Upon clarification that the experiment was noncommercial and purely for research purposes, they voluntarily chose to participate without compensation. Consequently, we contributed 10 yuan RMB to the pilgrimage sites associated with each Taoist participant. In contrast, the atheist participants received a stipend of 10 yuan RMB for their participation.
Demographic Information and Religious Behavior of Participants (Standard Deviations Are Presented Within Parentheses).
Materials
Following the methodologies employed by Hwang and Chun (2018) and Li (2023), we instructed participants to engage in a picture description task designed to assess how prior exposure to active and passive syntactic constructions influences subsequent language production. The syntactic priming effect was quantified by the proportion of passive sentences produced following the priming of both active and passive structures.
Participants were shown a series of images depicting actions with clear agents (doers) and patients (receivers of the action). Following the methodology of Hwang and Chun (2018), each of the 16 target images focused on transitive events that could be described using both active and passive voice:
Active voice: 狗 咬 了 老人
Gǒu yǎo le lǎorén
The dog bit a particle indicating a completed action the old man
Lǎorén bèi gǒu yǎo le
The old man passive voice marker The dog were bitten particle
Each trial comprised a prime image followed by a target image, both depicting actions that utilized the same verb. The experimental design employed two distinct lists of material, each consisting of image pairs where the prime image could be articulated using both active and passive constructions. An equal number of active and passive primes were included. The arrangement of these trials was dual-sequenced—both forward and reverse—to control for potential order effects and ensure that any observed priming effect was not merely attributable to the sequence in which stimuli were presented. The presentation of the primes was counterbalanced across subjects.
Across all participants, the positions of the agent and patient were systematically varied. For instance, in some trials, the agent might be on the left and the patient on the right, and in other trials, the opposite arrangement was used. By counterbalancing, researchers aimed to mitigate any potential confounding effects that could arise simply from the spatial location of the images. To prevent participants from becoming overly focused on the specific syntactic structures being tested (active vs. passive voice), the experiment also incorporated 16 filler trials. They depicted intransitive events (e.g., 女人在笑 “The woman is laughing”). These trials were similar in format to the target trials but differed critically in syntactic structure.
Procedure
To simulate real-life social interactions, we employed the confederate scripting technique, similar to approaches used in previously published studies on syntactic priming (Branigan et al., 2000; Li, 2023). In this task, the confederate acted as a genuine participant but operated under a predetermined script that guided their picture description activities in interactions with actual participants. By scripting the confederate's responses and actions, researchers can control variables in the experiment and facilitated a more structured exploration of syntactic priming effects.
We conducted this experiment at a community center located in a neighborhood with temples. This location was selected due to its accessibility for both religious and nonreligious participants. Participants were instructed to engage in a picture description task with a confederate online. They received a structured introduction to the experiment presented on a laptop, which informed them that this communication task entailed both picture matching and description. Both participants and the confederate took turns describing images to each other, while listeners verified the accuracy of the spoken descriptions. The participants were unaware that the confederate used fixed linguistic constructions, as defined by the predetermined script. This script specified whether to use an active or passive construction, which the confederate then read aloud as each image appeared on the screen. This mechanism ensured an active and focused examination of how different syntactic constructions influence participants’ grammatical choices.
Each trial commenced with a 500 ms fixation cross, followed by the presentation of a prime image (e.g., the chicken and the rabbit) alongside a confederate articulating the prime sentence, “The chicken bit the rabbit.” Participants were then required to assess the accuracy of the confederate's description: they pressed the “J” key if the description aligned with their perception of the image, and the “F” key if it did not. Following this, participants viewed a target image (e.g., the dog bit the old man) and were asked to describe it. Figure 1 illustrates the structure of an experimental trial, highlighting the relationship between the prime and target pairs.
An example of an experimental trial.
After participants completed one trial, the prime stimuli were followed by a target image, which they were then prompted to describe. When it was the confederate's turn to make a judgment, the target picture reverted to a prime picture for the next trial. Speech samples were collected throughout the task via a desktop microphone apparatus. Each audio file was orthographically transcribed by two pretrained coders to ensure both efficiency and intertranscriber consistency. Before commencing the main experiment, participants engaged in several practice trials to familiarize themselves with the procedures.
Following the completion of the experiments, a debriefing procedure was implemented to assess participants’ awareness of the true purpose of the task such as the link between Taoism and enhanced syntactic priming (e.g., “Now that we’ve completed the study, can you explain what you think the main purpose was?”). None of the participants demonstrated any awareness of the underlying experimental conditions. Subsequently, participants were fully informed about the actual objectives of the study. All of them expressed an understanding of the deceits employed during the study and provided informed consent for the utilization of their data for scientific purposes.
Coding and Analysis
Participant responses were first transcribed verbatim to ensure the accurate capture of linguistic structures which were then categorized as active voice, passive voice, or other forms. Specifically, an utterance is identified as typical active voice if it foregrounds an agent performing an action on a patient. For example, in the sentence “The cowboy hit the boxer,” the cowboy (agent) is performing the action (hit) on the boxer (patient). In contrast, a sentence is considered passive if it highlights the action and the patient, rather than the agent. This structure positions the patient as the subject and includes the passive marker “被 (bei)” to denote the passive voice. For example, “The boxer was hit by the cowboy” centers on the boxer (patient) receiving the action, with mention of the agent (cowboy) following later in the sentence.
Participants’ responses were considered correct as long as they provided an appropriate description of the event, even if their initial wording or structure was slightly altered. For example, if a participant said, “the bird, oh no, the fish was eaten by the bird,” the response was still considered valid because it describes the event effectively, despite the initial reference to “the bird.” The use of redundant words, such as adjective (e.g., “a large horse”), did not affect the validity of the response. Thus, additional words that did not change the core meaning of the description were acceptable. Responses were excluded from the primary analysis under two specific conditions: (a) Lack of action verb: If the utterance did not include an action verb, it was not counted. For example, simply stating “the girl and the boy” lacks an action and would be excluded; (b) word substitution: If participants changed key words from the original prompts (e.g., substituting “policeman” with “fireman”), their response was also excluded because it altered the intended meaning of the event being described. Overall, about 3% of the trials were excluded from the analysis.
Results and Discussion
We utilized the proportion of passive sentences produced when both active and passive structures were primed as the dependent variable to measure syntactic priming effects (Figure 2). To investigate differences in syntactic priming effects between Taoists and atheists, we conducted a mixed-model ANOVA, with Group (Taoists vs. atheists) as the between-subjects factor and Prime Condition (active vs. passive) as the within-subjects factor. There was no significant main effect of Group, F (1, 75) = .53, p = .47. However, we observed a significant main effect of Prime Condition, F (1, 75) = 552.30, p < .001, ηp2 = .88. Specifically, the proportions of passive sentences produced were 7.47% when the confederate generated active sentences, compared to 35.06% when the confederate generated passive sentences.
Priming effects in percent when the active and passive constructions were primed in atheists and Taoists. Error bars show standard errors of the mean.
Importantly, the interaction between group and prime condition was significant, F (1, 75) = 12.49, p < .001, ηp2 = .14. Analysis of this interaction using simple effects revealed that the proportion of passive sentences was higher when participants were primed with passive structures than being priming with active structures for both atheists (p < . 001) and Taoists (p < . 001). This suggests that the syntactic priming effects can be observed in both cultural groups. Additionally, Taoists tended to produce a higher proportion of passive sentences than their atheist counterparts in passive priming conditions (p = .01). Thus, these findings support our main hypothesis that syntactic priming effects are more detectable among Taoists than among atheists.
Although Study 1 provided preliminary evidence supporting our hypothesis that Taoism was associated with enhanced syntactic priming effects, some critics might suggest that the observed relationship could stem from differences in religiosity between Taoists and atheists. In other words, an alternative explanation might be that all religious individuals are more inclined to align syntactically with others, thus the observed effect might not relate specifically to the philosophical tenets of Taoism, namely, moving with the flow.
To address this potential confound, we conducted Study 2, which involved both Taoists and Buddhists matched for levels of religiosity. If differences in the magnitude of syntactic priming effects are still apparent between these two groups, it would bolster our confidence that it is the specific beliefs of Taoism, rather than religiosity in general, that contribute to stronger priming effects.
Study 2
Method
Participants
The research was conducted in Wuhan, the capital of Hubei Province in central China. Employing G*Power for statistical analysis, we determined that 48 participants were necessary to detect a medium effect size observed in Study 1 (ηp2 = .11) in a two-way repeated measures ANOVA with a power of 0.95. However, a knowledgeable reviewer suggested that we reevaluate our analysis concerning the priming effects and with the right effect size for the interactive effect ηp2 = .14, 20 participants were required to reach 95% statistical power. Either way, to accommodate potential participant attrition and to ensure robust statistical analyses, we increased the number of participants. A total of 49 Taoists were recruited as these numbers were feasible within the constraints of our time and funding. A comparably sized group of Buddhists (N = 49) was also included. These participants were carefully matched with their Taoist counterparts across various sociodemographic variables including age, gender, geographical location, and educational levels (for more details, see Table 2).
Demographic Information and Religious Behavior of Participants (Standard Deviations Are Presented Within Parentheses).
All Taoist and Buddhist monks had been adherents of their respective religions for no less than 6 years at the time of the study. They were recruited from local temples, where they were primarily engaged in community welfare rituals. The participants resided in monasteries maintaining intense religious practices and beliefs. All participants declined remuneration, citing that acceptance would contravene specific ethical or doctrinal standards. Upon clarification that the experiment was noncommercial and purely for research purposes, they voluntarily chose to participate without compensation. Consequently, we contributed 10 yuan RMB to the pilgrimage sites associated with each religious participant.
Materials and Procedure
All materials and procedure were the same as Study 1.
Results and Discussion
The coding and analysis were identical to those used in Study 1. Overall, fewer than 3%of the trials were excluded from the analysis. We utilized the proportion of passive sentences produced when both active and passive structures were primed as the dependent variable to measure syntactic priming effects (Figure 3). To investigate differences in syntactic priming effects between Taoists and Buddhists, we conducted a mixed-model ANOVA, with Group (Taoists vs. Buddhists) as the between-subjects factor and Prime Condition (active vs. passive) as the within-subjects factor. There was no significant main effect of Group, F (1, 96) = .06, p = .81. However, we observed a significant main effect of Prime Condition, F (1, 96) = 692.05, p < .001, ηp2 = .88. Specifically, the proportions of passive sentences produced were 8.80% when the confederate generated active sentences, compared to 36.35% when the confederate generated passive sentences.
Priming effects in percent when the active and passive constructions were primed in Buddhists and Taoists. Error bars show standard errors of the mean.
Importantly, the interaction between group and prime condition was significant, F (1, 96) = 8.54, p = .004, ηp2 = .08. Analysis of this interaction using simple effects revealed that the proportion of passive sentences was higher when participants were primed with passive structures than being priming with active structures for both Buddhists (p < . 001) and Taoists (p < . 001). This suggests that the syntactic priming effects can be observed in both cultural groups. Additionally, Taoists tended to produce a higher proportion of passive sentences than their Buddhists counterparts in passive priming conditions (p = .04). Thus, these findings support our main hypothesis that syntactic priming effects are more detectable among Taoists than among Buddhists.
General Discussion
While the philosophy of “moving with the flow” in Taoism advocates for aligning with the natural order of the universe rather than resisting it, the manner in which this accommodation and adaptability are manifested in interpersonal communication, particularly linguistic behavior, remains unclear. This research explores this issue by examining the link between Taoism and enhanced syntactic priming effects. In concordance with our hypothesis, Study 1 found that compared to atheists, Taoists exhibited a greater preference for repeating the syntactic forms used by their interlocutors. In Study 2, we replicated the findings of Study 1 by demonstrating that Taoists were more likely to adopt the same syntactic structure when interacting with a specific interlocutor than Buddhists did. Overall, these findings suggest that the Taoist concept of “moving with flow,” rather than general religiosity, significantly influences syntactic alignment processes.
Note that the priming effects observed in our research are relatively large. One possible explanation for this phenomenon is likely the lexical overlap. Specifically, the repetition of a particular lexical item, such as verbs in this case, can enhance the likelihood of employing the same syntactic structure in subsequent sentences. For instance, Traxler et al. (2014) found that priming effects during online comprehension were significantly strengthened when the same verb was repeated in both the prime and target sentences. This finding underscores the role of lexical items in the activation and retrieval of corresponding syntactic structures during language processing.
However, since both groups of participants in each study completed identical syntactic priming tasks, this suggests that the facilitating effect of same-verb priming should be comparable across conditions and groups. Thus, the lexical boost effect should not serve as a confounding factor in our research. Nevertheless, the observed syntactic priming effects may be less pronounced in tasks that lack verb repetition. Future research should investigate whether our findings can be generalized to syntactic priming scenarios that do not involve verb repetition.
Cognitive scientists have traditionally posited that syntactic priming is primarily an autonomous process, relatively immune to nonlinguistic factors (Dell, 1986; Pickering & Branigan, 1998). This presumed automaticity suggests that our brains utilize past linguistic experiences to streamline communication, reflecting how language processing is often more instinctive than deliberate. For instance, Branigan et al. (2003) observed that participants exhibited a strong tendency to repeat the syntactic form of their interlocutor's immediately preceding utterance, regardless of whether they perceived the partner to be a human or an unintelligent computer. Thus, these observations support the notion that structural priming is a largely automatic cognitive process, unaffected by the mental states of the interlocutor.
However, an increasing body of literature challenges this view by demonstrating that syntactic priming is not entirely independent but is influenced by sociocognitive factors. Utilizing virtual reality techniques, Heyselaar et al. (2017) investigated how social perception impacts syntactic processing. The researchers found that participants displayed the most substantial priming effects with moderately strange avatars, with effects diminishing in interactions with the least or most strange avatars. These findings suggest that social perception of interlocutors may modulate priming magnitude. Our work enriches this stream of literature by showing that religious systems can exert additional influence on how interlocutors linguistically align with their conversational partners.
The Taoist philosophy of “moving with flow,” which advocates for aligning oneself with the natural order of the world, suggests that the most effective approach to handling situations is not through direct action, manipulation, or force, but rather through a form of intelligent spontaneity that harmonizes with the natural progression of events (Simpkins & Simpkins, 2015; Wang & Stringer, 2000). Applying this philosophy in communicative contexts may indicate that Taoists are more attuned to the context and dynamics of interactions. In line with this perspective, we found that syntactic priming effects were more detectable among Taoists than among atheists and Buddhists. Therefore, our findings suggest that the observed relationship is best explained by the joint effects of recent syntactic exposure and their underlying philosophy of accommodating and adapting through “moving with the flow.”
As the discipline of language science evolves, it grapples with fundamental questions concerning the generalizability of empirical findings like many domains within the social sciences (Grieve, 2021). A promising approach to address these concerns is the diversification of samples, populations, contexts, and epistemologies. Recent studies suggest that individuals from non-Abrahamic religions exhibit divergent patterns of language and cognitive processes compared to Western, Educated, Industrialized, Rich, and Democratic populations (Guan et al., 2024; Li & Cao, 2019, 2021). For example, Li and Cao (2018) observed that Buddhists and Taoists employ distinct types of spatiotemporal metaphors in their mental models. Specifically, Buddhists typically adhere to a “past-in-front” mapping, whereas Taoists prefer a “future-in-front” approach. These findings underscore the importance of exploring varied sociocultural ecologies for theoretical and empirical advancements in language science.
The current research adds data from an understudied population, including both Taoist and Buddhists monks, which differ significantly from traditional samples that are disproportionately drawn from geographically and socioculturally narrow cohorts. By incorporating these diverse groups, our study offers a more comprehensive understanding of the scalability of syntactic priming effects and the factors influencing observed effects. The commitment to diversity, equity, and inclusion is not solely exemplified by sampling from heterogeneous populations but is also mirrored in the sourcing of diverse theoretical frameworks. Notably, while many studies outside the Western context merely attempt to replicate Western-origin theories, our theoretical framework is derived from observations of Taoist doctrines. Given the strong parallels between the Taoist concept of “moving with the flow” and CAT, our research illustrates how linguistic and cognitive sciences can benefit from incorporating diverse theoretical perspectives from other cultures, such as religion and philosophy.
In addition to diversification in sampling, another methodological strength of our research lies in the adoption of the “just minimal difference” approach (Uskul et al., 2008). This method focuses on subtle and often underexamined factors that may impact cross-cultural variations. For instance, in Study 2, we minimized potential confounding variables between Taoists and Buddhists, thus isolating the culturally specific concept of “moving with flow” inherent to Taoism.
This methodology bolsters confidence in the specific influences of religious beliefs on syntactic priming. It is important to note, however, that the design was quasi-experimental. It is not feasible to randomly assign participants as Taoists, Buddhists, or atheists—participants had already identified as Taoists, Buddhists, or atheists prior to their involvement in our study. Thus, a more rigorous test for causal inference would involve longitudinal data to examine whether practitioners of faith exhibit varying degrees of syntactic priming before and after undergoing religious conversions or deconversions.
In addition to its theoretical and methodological contributions, this research acknowledges several limitations that warrant future investigation. First, although we discovered that religious systems moderate the magnitude of syntactic priming, the mediating mechanisms underlying this effect remain unclear. One hypothesis is that differences in attention and memory encoding may explain these effects (Kruschke, 1992, 2011). For instance, Taoism promotes the belief that one's actions impact others, thereby encouraging adherents to cultivate harmonious interpersonal relationships. If Taoists place substantial emphasis on the dynamics of interpersonal interactions and consequently focus more intently on their conversational partners and their speech, this could lead to enhanced encoding of sentence structures, resulting in greater priming. To elucidate the relationships presented here, future research should explore the mediating roles of attention and memory.
Second, the current research utilized a picture-matching and -describing game to measure syntactic persistence in a laboratory setting. While this approach is valid, it may face criticism for its lack of ecological validity. Critics may argue that if the grammatical choices observed in this task do not reflect actual linguistic behaviors in real-life contexts, the findings’ significance and interest could be diminished. To mitigate this issue, future research could employ multiple methodologies and data sources to enhance both the validity and reliability of the findings. For example, combining qualitative approaches (such as interviews or discourse analysis) with quantitative methods (such as psycholinguistic experiments or statistical analysis) could offer a richer examination of language use within the Taoist community.
Third, the current standard analytical approach for this type of experiment is to utilize mixed logistic regression analyses on response frequencies rather than ANOVA on proportions in the present research. This approach was deliberately chosen to ensure consistency with methodologies employed in recently published papers in this journal (Li, 2023), thereby facilitating comparability with previous research. Furthermore, to mitigate researcher degrees of freedom, we made a priori decisions regarding the analyses to conduct, although this approach was not formally preregistered. Additionally, the sample sizes in both studies were determined based on this ANOVA test. Future research should employ mixed logistic regression analyses to avoid known issues associated with analyzing proportions, which may be influenced by varying response counts across conditions, as well as to accommodate both participant- and item-level variability.
Finally, the dataset for the present study is based on a relatively small number of participants, consisting exclusively of males. This is not surprising given that Taoist and Buddhist monks comprise a very hard-to-reach population. Despite this practical constraint on participant recruitment, the small sample size carries a heightened risk of Type I and Type II errors. To mitigate this, we conducted a prior power analysis to determine the requisite sample size and predefined our statistical analyses before commencing the experiment. Nonetheless, future research should aim to replicate the current findings in a larger population of Taoist practitioners in China and other countries to ascertain the robustness of our findings.
Conclusion
Religion pervades daily life as it significantly influences various aspects of individual behavior and social practices. However, religion is a heterogeneous concept, encompassing a wide variety of beliefs, practices, traditions, and organizational structures that vary markedly across different cultures and communities. The current research examined how Taoist beliefs influence the syntactic choice behavior of their adherents. We found that Taoists showed stronger priming effects in interpersonal communication compared to atheists and Buddhists. This effect cannot be solely attributed to differences in religiosity, as Taoists also demonstrated a more detectable syntactic priming effect than Buddhists with comparable levels of religiosity. One important conclusion of our study is that the influence of different religious affiliations on linguistic behavior is much greater than commonly perceived, owing to their fundamentally distinct beliefs. Ultimately, this highlights the profound impact of religious identity on cognitive and communicative processes.
Footnotes
Acknowledgments
The feedback from the anonymous reviewers and the editor on prior drafts of this manuscript were helpful and much appreciated.
Declaration of Conflicting Interests
The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Sichuan International Studies University 2024 Special Project, Second Round of Chongqing’s First-Class Disciplines in Foreign Languages and Literature 2023 Research Projects, Research Project of Humanities and Social Sciences of the Ministry of Education (Grant Nos. SISU202423, SISUWYJY202302, and 24YJA740022).
