Introduction

National research programs are developed to promote innovative development and growth within specific domains, aiming to address intricate, interdisciplinary, and multifaceted scientific issues. The scientific research project teams in China can be classified into three levels: those financed by national special funds, those supported by the Chinese Academy of Sciences, the National Defence Science and Industry Commission, and provinces and municipalities directly under the Central Government; those supported by some universities, research institutes, and enterprises; and those financed by some universities, research institutes, and enterprises on their own. RTI is a complex, high-risk, and lengthy process that cannot be achieved without a combination of internal and external factors, such as financial support, talent pool, effective team management, favorable economic environment, and policy support.

NRPY typically undertake major national-level scientific research tasks, involving complex, interdisciplinary, and multi-level scientific issues. Compared to other types of organizations, these teams have a more centralized decision-making process in their innovation activities, tighter resource allocation, and face higher demands for technological innovation. This makes them uniquely important in achieving RTI. Relevant scholars have examined the significant positive impact of IPP on the technological innovation within NRPT1. Nonetheless, some scholars argue that strict IPP may negatively affect technological innovation in national research projects2. Therefore, further exploration is necessary to understand the impact of IPP on RTI. Currently, there is a lack of literature examining the role of IPP from the internal perspectives. IM refers to the creation of a cultural Milieu within an organization or team that actively encourages innovative thinking, respects diverse perspectives, and advocates open communication. In addition, the IM also focuses on tolerating failure and accepting feedback, so that team members are not penalized for failure, but rather learn from it to continuously refine and improve their innovative methods and ideas. Although other internal factors such as team management, resource allocation, and personnel incentives may also have a significant impact on technological innovation3, the innovation environment is particularly emphasized from an internal perspective because it can shape the innovative behavior of team members and enhance the overall innovation capacity of the team4. According to innovation-driven theory, the innovation environment not only stimulates individual creativity but also fosters knowledge sharing and collaboration within the team, which is especially crucial for complex national research projects. Therefore, the innovation environment plays a pivotal role in promoting technological innovation. GP, on the other hand, represents the collective intelligence and cooperative ability of team members to achieve innovative results beyond individual capabilities through effective collaboration and communication. Cooperation and collaboration among team members can integrate their unique knowledge and skills to form more innovative and comprehensive solutions. By sharing knowledge, experience and resources, teams are able to respond quickly to market needs and challenges, and efficiently promote the implementation of innovative projects. When the IM and GP are fully combined, the team will usher in greater innovation breakthroughs and success. In such a team, each member is fully stimulated by the passion for innovation and dares to take risks to try new methods and ideas. The close cooperation and open communication among them promote the collision and integration of different views and ideas, which provides a constant source of motivation and inspiration for innovation. Therefore, by establishing a positive IM and giving full play to the potential of the group, the team can truly unleash its potential and achieve RTI. In conclusion, exploring whether IPP can affect the RTI of NRPT, and the chain mediating effect of IM and GP will be helpful to improve the management effectiveness of research teams in universities and enterprises.

This study conducted field research to obtain a matched dataset and employed PLS-SEM and ANN models for empirical analysis. According to the PLS-SEM results, moderate strengthening of IPP may contribute to the realization of RTI within NRPY. Through ANN- conducted sensitivity analyses, IPP has been identified as the most crucial predictor.

The primary contributions of this research can be condensed into three key areas:

  1. ①

    Expanding two fundamental theories: the innovation-driven theory and the AMO theory. These theories enhance the existing theoretical framework for Research and Innovation Technology-based (RTI-type) studies on IPP and NRPT. Research has been conducted to explore the influence of external factors on the RTI of NRPT, with a focus on IPP5, instead, this paper highlights the importance of internal team elements, such as IM and GP, in driving technological advancements. This study delves into the RTI, a subject that has not been extensively examined in the context of technological innovation, thereby enriching the existing literature in this field.

  2. ②

    The objective of this passage is to enhance the research framework for national research project team management. Existing research predominantly focuses on single latitude analysis6, this study analyzes from multiple dimensions. By combining linear and nonlinear analyses, this study further elucidates the causal relationships underlying RTI within teams.

  3. ③

    This paper discusses the policy implications and recommends that countries establish a robust intellectual property policy to protect the rights and interests of scientific research outcomes. Furthermore, the development of policies that foster collaboration between scientific research teams and industry for technology transfer is crucial. Governments and organizations can enhance their leaders’ integrated capabilities through effective training and development programs. Moreover, providing increased financial support for scientific research projects can lead to increased R&D investments, thereby increasing the likelihood of RTI within these research teams.

Theoretical frameworks and hypotheses

To better uncover the mechanisms by which IPP influences RTI in national research projects, this study employs Michael Eugene Porter’s innovation-driven theory and the Ability-Motivation-Opportunity (AMO) theory proposed by Appelbaum to explore how IPP fosters RTI within research teams7,8,9. The innovation-driven theory posits that a nation’s economic development progresses through four stages: factor-driven, investment-driven, innovation-driven, and wealth-driven. Innovation-driven development emphasizes technology, knowledge, institutions, and management as key drivers, making resource allocation more efficient and rational through the application of new knowledge and technology. This process continuously enhances the quality of the workforce and improves the research environment, thereby elevating innovation levels. IPP similarly promotes productivity by stimulating technological innovation within research teams, improving investment efficiency, and enhancing competitive advantage10. Barrios et al. note that the relationship between open innovation and confidentiality agreements, which serve as tools for protecting intellectual property, may influence the innovation performance of manufacturing and service companies11. The AMO theory suggests that if the ability, motivation, and opportunity requirements of team members are met, the benefits to the team will be maximized. These two theoretical frameworks are selected because they provide structured approaches to understanding the complexities of research innovation, particularly in elucidating the specific impacts of institutional factors and team dynamics on the innovation process. First, the innovation-driven theory highlights the central role of knowledge and technological innovation in modern economic development and underscores the importance of institutional environments, such as IPP, in maintaining this innovative momentum12. This theory offers a macro-level perspective for assessing how institutions shape the environment and incentive structures for technological innovation. Additionally, the AMO theory provides a micro-level perspective by analyzing how the interplay of ability, motivation, and opportunity at the team level affects organizational performance. This theory offers a practical framework for understanding and influencing the dynamics within teams13.

This study aims to address a crucial factor often overlooked in current research—namely, while IPP is widely recognized as a driver of innovation, existing studies often fail to thoroughly explore how it specifically impacts the innovation process through the practical operations of research teams. Specifically, this paper utilizes the analytical framework of the AMO theory to investigate how IPP fosters high-quality technological innovation outcomes by enhancing researchers’ motivation (e.g., through patent incentives and increased R&D investment), strengthening their abilities (e.g., through training and technical support to boost group potential), and creating opportunities (e.g., by promoting interdisciplinary collaboration, enhancing integrative leadership, and optimizing the innovation atmosphere). For example, IPP enhances researchers’ motivation to pursue high-risk innovation projects by ensuring the commercial returns of their research outcomes. It also strengthens research teams’ capabilities to access and utilize critical technological resources through institutional safeguards14. Additionally, by promoting interdisciplinary collaboration, IPP creates new opportunities for team members to engage in exchange and cooperation15. This theoretical integration not only deepens our understanding of the mechanisms by which IPP operates within research teams but also provides concrete recommendations for improving research management and policy-making, thereby enhancing the study’s theoretical depth and practical relevance. By conducting a detailed analysis of the multidimensional impacts of IPP, this study supplements existing literature and provides a scientific basis for the formulation of more effective innovation-promoting policies.

The success of national research programs in achieving RTI and their impact on technological innovation are primarily influenced by R&D investments and the stock of intellectual capital. A higher R&D investment creates a more favourable environment for promoting technological innovation and progress. Drawing on Clayton Christensen’s The Innovator’s Dilemma, many organizations are drawn towards maximizing profits and meeting current demands, yet overlook the exploration and development of innovations in niche technology tracks16. RTI enables the transformation of knowledge and technology into innovative outcomes within national research programs, fostering and expanding the boundaries of the discipline and its applications across various fields. Consequently, proactive advancements in technological innovation are crucial for national research programs to prevent losses in science and technology competitiveness, not just for individual enterprises but also for the nation as a whole. Furthermore, a robust inventory of knowledge assets serves to bolster technological innovation capabilities. A substantial knowledge asset inventory will foster a stronger technological foundation for innovation teams working on national research projects. A shortage of such assets could hinder the team’s progress, establish a lower starting point for technological innovations, and heighten the challenges in achieving breakthroughs. The theory of knowledge creation suggests that the sharing of tacit knowledge within an organization, combined with the processes of socialization, externalization, integration, and internalization, can lead to knowledge innovation17. For instance, a discord among team members, like in an innovation team, might cause the inadequate utilization of knowledge assets, impeding the achievement of RTI in national research projects18.

The full utilization of knowledge assets and substantial R&D investments is crucial for enhancing the realization of RTI. Nevertheless, pure market forces may demonstrate some market failures19. It was recognized that additional policy measures are required to enhance innovation capacity. The government-promoted IPP system plays a vital role in fostering innovation improvement. By effectively protecting proprietary rights and reducing technological innovation uncertainties, IPP contribute to achieving RTI in national research projects20. The IPP can implicitly influence innovation team leaders by mitigating their apprehensions about innovation disclosure, thereby encouraging them to adopt more proactive technological innovation behaviors. Inversely, either insufficient or exaggerated IPP safeguards in research projects can significantly affect R&DII21, and in turn, impact RTI. The spillover effect tends to be weak when IPP awareness is heightened, while the complexity of the technology makes it difficult to replicate knowledge assets during the transfer process. As a result, it’s more plausible for imitators to avoid infringing on the originator22, thus fostering RTI. Based on this, this paper proposes:

H1

IPP mitigates spillover effects, significantly enhancing the radical innovation capabilities of national research project teams.

The strength of IPP may, to some extent, boost team IM within national research projects. In particular, stringent IPP laws serve to deter plagiarism, and IP licensing provides a means to utilize others’ patented technologies, which is more conducive to the development of good team IM. This primarily stems from the theory of catalytic disclosure, which posits that the sharing and disclosure of information can lead to more innovative ideas and discoveries. Nonetheless, without an appropriate IP framework, individuals might not invest their time and resources into this behavior. A strong IPP can mitigate this risk, promoting collaboration and co-creation between academics and businesses23.

In national research projects, a robust IM culture fosters transparency in information sharing, thus increasing the likelihood of achieving RTI. Undoubtedly, a corporation with robust innovative management capabilities can encourage a proactive mindset towards exploring novel ideas and perspectives, while simultaneously providing the required resources and financial support for the implementation of technological innovations and transformation strategies. The primary reason for this is that IM can help teams overcome the challenges posed by emerging technologies such as uncertainty and complexity24. Furthermore, the team’s visionary, adaptive, and communicative skills, which are crucial attributes influencing IM25, serve as a solid foundation for NRPT to foster creativity and achieve RTI.

The impact of IPP on the achievement of RTI by NRPT is indirect, with the team’s IM playing a pivotal mediating role in this process. NRPT and members are more likely to perceive IPP as a driving force for technological innovation rather than an obstacle when working in an IM system that emphasizes high risk tolerance, openness to new ideas, and a propensity for collaboration and learning26. In a strong IPP environment, team members’ concerns about intellectual property are reduced, which lowers the risk associated with information sharing, thereby facilitating the free flow of knowledge and increasing the willingness to collaborate. Open communication and active cooperation within an innovative atmosphere are essential foundations for generating new ideas and achieving innovative breakthroughs within teams27. In this process, IPP indirectly fosters innovative behavior by creating an environment of trust. In a strong IPP environment, the innovative atmosphere also enhances the team’s tolerance for risk, allowing members to make mistakes during experimentation and innovation without facing penalties28. This inclusive innovation culture encourages more attempts and experimental innovations, thereby increasing the likelihood of successfully achieving RTI. By alleviating concerns about failure, IPP promotes more adventurous and risk-taking innovative behaviors. The theory, based on the “knowledge spillover theory,” asserts that intellectual property rights (IPRs) impede the transfer and dissemination of knowledge, leading to sub-optimal outcomes for society. This theory, however, overlooks the complexity of the IP protection system and the varying degrees of exclusivity provided by IPRs, as well as the intermediate factors that impact the influence of IPRs on various mediating factors during innovation, such as team IM, R&D expenditures, human resource allocation, partnerships, and knowledge management practices29,30. The interplay between team knowledge integration (IM) and IPP is examined. The results indicate that increased IPP levels positively impact IM within NRPT, especially in collaborative and learning-oriented environments. Based on this, this paper proposes:

H2a

Intellectual property protection positively affects the innovation milieu of national research project teams;

H2b

National research project team innovation milieu has a mediating effect between intellectual property protection and radical technological innovation;

Under the protection of the intellectual property system, team members’ motivation for innovation is significantly enhanced. Strong IPP ensure that the creative work of team members is duly recognized and protected, encouraging them to invest more time and effort in exploring new fields and experimenting with new methods31. This heightened innovation enthusiasm and initiative not only amplify individual contributions but also, through synergy, unlock the collective potential of the group. This, in turn, enables the team to demonstrate greater creativity and execution capabilities when tackling complex technological challenges. IPP encourages members to openly share their innovative ideas and research outcomes, knowing that these results will be safeguarded within a legal framework32. This free flow of knowledge and sharing of resources is a critical factor in boosting collective potential. It fosters stronger collaboration within the team, ultimately enhancing the overall efficiency of innovation efforts.

The success of national research projects is influenced by the team’s global profile (GP). Higher GP levels are beneficial for IPP to affect RTI. Specifically, an increase in team GP positively correlates with patent generation and citations33,34, thus increasing the probability of RTI achievement by the NRP team. Diverse team members, possessing varying skills, experiences, and cognitive approaches, can provide a comprehensive range of perspectives and solutions for innovation. Integrative leaders, who acknowledge and effectively integrate these diverse perspectives, can stimulate innovative possibilities. IPP provides a secure environment for innovation, ensuring that the knowledge and innovations of team members are not easily misappropriated. In such an environment, trust among team members increases, and cohesion is strengthened. This enhanced GP, achieved by raising the level of collaboration among members, leads to closer cooperation during the technological innovation process, thereby boosting the overall innovative capacity of the team35. In a strong IPP environment, the protection of team members’ innovative achievements greatly stimulates their motivation for innovation and commitment to the team’s goals36. GP is not only reflected in the individual abilities of team members but also in the collective innovative synergy of the team37. When team members feel that their intellectual property is secure, they are more willing to invest additional effort and resources to advance innovation projects, driving the realization of RTI. IPP reduces the external disruptions and internal conflicts that teams may encounter during the innovation process, allowing them to focus more intently on tackling complex technological challenges14. In this process, GP plays a critical role by integrating the knowledge, skills, and experiences of team members, thereby enhancing the team’s overall ability to meet innovation challenges. Moreover, team members demonstrating adaptability and flexibility can aid the entire team in better coping with the uncertainties inherent in the innovation process. High-potential team members are more likely to be motivated to innovate, and they exhibit heightened commitment in high-stakes R&D environments. This translated into increased effort, which in turn increases the likelihood of innovation. Moreover, these individuals can allocate resources more effectively, optimize their use, and thus enhance the efficiency of R&D, accelerating the innovation process. Conversely, the GP of a national research project team has emerged as an essential metric for assessing team performance, regardless of the potential inhibitory or promotional effects of IPP on technological innovation38,39. GP in NRPT was linked to improved team coordination, communication, and task accomplishment40, which in turn led to higher technological innovation outcomes. This, in turn, facilitated teams to attain RTI. Based on this, this paper proposes:

H3a

Intellectual property protection positively affects the group potential of national research project teams;

H3b

National scientific research project team group potential has a mediating effect between intellectual property protection and radical technological innovation.

The impact of IPP on RTI is not only direct but also mediated through complex internal mechanisms. These mechanisms include the IM and GP, which together create a chain mediation effect that amplifies or modifies the influence of IPP on technological innovation. IPP provides research teams with a secure and protected environment for innovation, enhancing their trust and commitment to the team, thereby fostering a positive IM41. In a robust innovation atmosphere, collaboration and communication among team members are more seamless, and the level of trust is higher42. This innovation atmosphere provides the necessary foundation for the expression of GP, as members are more willing to share their knowledge and skills in a secure environment and work together to solve challenges in innovation. The innovation atmosphere encourages team members to continually push the boundaries of their individual and collective capabilities, resulting in a powerful collective intelligence and collaborative ability. As the innovation atmosphere deepens, GP is further stimulated and reinforced, enabling the team to more effectively integrate resources and tackle complex technological challenges. The innovation atmosphere serves as an initial feedback mechanism of IPP, enhancing team members’ willingness to collaborate and their capacity for innovation43. These factors, in turn, translate into substantial innovative outcomes through the amplification of GP. Therefore, the impact of IPP on RTI is not a simple linear relationship; rather, it is gradually amplified through a series of complex internal mechanisms, forming an interactive chain across various stages. Based on this, the paper proposes:

H4

The innovation Milieu and group potential of national scientific research project teams have a chain mediating effect between intellectual property protection and radical technological innovation.

Combining hypotheses H1 to H4, the theoretical model of the study was plotted (see Fig. 1):

Fig. 1
figure 1

Conceptual model.

Full size image

Research methods

Sampling and data collection

In this study, the members of the NRPT in Chinese universities and enterprises were investigated to assess the proposed hypotheses. The data were primarily obtained from eminent universities and research organizations, such as the University of Science and Technology of China, Harbin Institute of Technology, Harbin Engineering University, Anhui University, Northeast Agricultural University, and prominent high-tech companies. Furthermore, the selection process of NRPT involves multiple considerations, typically encompassing a diverse array of expertise from various sectors, including industry, academia, and government agencies. However, the involvement of multiple entities may lead to concerns regarding intellectual property rights, ownership, disclosure, and confidentiality. Numerous national research projects are significant financial investments funded by governments, funding bodies, academic institutions, corporations, and other stakeholders. The protection of intellectual property rights, including patents and technological achievements, against theft and infringement is of paramount importance.

This study aims to examine the effect of IPP on the RTI of NRPT, which is inhibited by the absence of secondary data relevant to these teams. Therefore, a questionnaire survey was conducted to gather sample data. A sample of 15 participants, consisting of 5 professors and 10 master’s and doctoral students, was initiallypre-tested to validate the questionnaire’s reliability. Subsequently, the questionnaire items were slightly adjusted based on the feedback gathered from this small-sample pilot survey. The research data was gathered from participants in national studies carried out by universities, research institutes, and corporations. A clear explanatory letter was employed to communicate the study’s purpose, emphasizing the voluntary nature of participation. This paper also guarantees to participants that their responses will remain anonymous and will only be utilized for academic purposes. The data for this study were gathered between October 2023 and November 2023. A total of 435 research questionnaires were disseminated, resulting in the receipt of 336 complete and valid questionnaires, with a validity rate of 77.2%. Table 1 presents the demographic characteristics of the study participants and the project team. It is clear that the selected sample is representative, guaranteeing the reliability, scope, and universality of the data source.

Table 1 Demographic characteristics statistics (N = 336).
Full size table

This study was approved by the ethical review board of East University of Heilongjiang (Protocol Identification Number KY2023-01) and was conducted in accordance with the principles specified in the Declaration of Helsinki. Informed consent was obtained all from all subjects and/or their legal guardian(s) to participate in publishing the information and images in an online open-access publication.

Measurement instruments

In the current research, four questionnaires were utilized to examine the proposed hypotheses. The indicators within each variable’s scale were chosen or modified from existing scales, in line with the study’s requirements. The scale was constructed using 53 first-order observation entries. Descriptive statistics and correlation analyses were conducted on the variables in Table 2, which encompassed four variables: IPP, RTI, IM, GP. The reliability and validity of the questionnaire scales in this study were evaluated through pretest and standardized measures of reliability and validity. Unlike the demographic section, all survey items were rated by respondents on a 5-point Likert scale, ranging from 1 (very low) to 5 (very high), thus encompassing various study variables. Participants provided detailed demographic information, including gender, age, educational background, and project participation duration, along with other relevant characteristics.

Table 2 Descriptive statistics and correlation analysis (N = 336)
Full size table
  1. 1.

    Intellectual property protection. In accordance with Xiao’s framework, we selected three levels of the scale, totaling 19 entries, to create an assessment scale that is consistent with the capability to protect intellectual property rights44. The internal consistency test for the three latent dimensions of the scale was carried out independently, resulting in Cronbach α coefficients of 0.953 for IPR protection mechanism, 0.940 for IPR protection strength, and 0.941 for IPR protection timeliness.

  2. 2.

    Radical technological innovation. By combining the research of Wang et al. and Wang et al., the capacity for RTI in national scientific research project teams was assessed across five dimensions: academic output, talent output, result transformation, economic output, and social benefit45,46. These dimensions comprised a total of 21 entries. The internal consistency test for the five dimensions of the scale was independently conducted, resulting in Cronbach α coefficients of 0.909, 0.916, 0.943, 0.918, and 0.922, respectively.

  3. 3.

    Innovation milieu. Furthermore, the IM is described according to Liu et al.‘s scale47. The development of an assessment scale congruent with the innovative environment of the national research project entailed selecting two dimensions of the scale consisting of six items, which were colleague support and leadership support. The internal consistency assessments for the two latitudes of the scale were carried out separately, resulting in Cronbach alpha coefficients of 0.769 for the colleague support and 0.908 for the leadership support.

  4. 4.

    Group potential. Adopting Guzzo’s scale, four dimensions were chosen: innovation culture, strategic foresight, learning ability, and sustainable benefits48. These formed an assessment scale to evaluate GP, in line with the national scientific research project team. Internal consistency testing was carried out, and the Cronbach α coefficient was calculated to be 0.948.

Common method bias

To reduce common method bias (CMB) during the questionnaire completion stage, this study adopted procedural adjustments for data collection, ensuring consistency across different respondents and adhering to Podsakoff et al.‘s recommendations49. The primary focus of the text is to provide guidelines for answering questions and fostering a deeper understanding of specific concepts. To ensure the confidentiality, anonymity, and voluntary participation of the respondents, the researcher highlighted the significance of truthful responses. Moreover, it was communicated to the participants that there were no preconceived correct or incorrect answers. Subsequently, a meticulous examination of each item was performed to verify the absence of obscure, misleading, or uncommon terms. The wording was further simplified to enhance comprehension, and the statement order was adjusted to minimize the possibility of respondent “guessing”50. This approach ensures the rationality and conciseness of the observations presented in this study.

Moreover, this study Utilized Harman’s one-way test and Lindell & Whitney’s labelled variable approach for post-hoc examination51,52. Before attempting the hypothesis tests for the research model, Harman’s one-way test assessed whether there was a common method bias in the questionnaire data. The analysis reveals that the highest factor-explained variance was 20.854%, below the recommended 40%. However, the total explained variance of all extracted factors amounted to 72.26%, exceeding the suggested threshold level by 50%. Consequently, common method bias is not an issue for this dataset, as confirmed by several recent studies53. Subsequently, Lindell and Whitney’s approach was adopted to examine whether common method biases affect the reliability of study results52, utilizing an theoretically uncorrelated variable as a marker (marker variable). In the ongoing study, a marker variable (“collusion with other members and general interpersonal tension”) was used to assess the alignment of participants with their interpersonal styles. The analysis by Lindell and Whitney revealed minor and non-significant correlations (ρ = 0.01–0.04), indicating the absence of a common methodological bias in the study’s data52. This implies that the data in this study might have been free from common methodological biases.

Data analysis

To examine the model’s assumptions, the study utilized PLS-SEM to analyze the cross-sectional data from the NRP team, and SmartPLS 4.0.9.5 software was employed to evaluate these assumptions. Structural equations and baseline regression differ in their comprehensive analysis, involving multiple independent and predictor variables. Khan et al. suggest that a structural equation model demands a minimum of 100 sample data inputs for optimal accuracy and reliability54. This study’s valid sample data includes 336 entries, adhering to the suggested guidelines by Khan et al.54. Structural equation modeling offers a comprehensive analysis of the mean, variance, and covariance for each variable, simultaneously generating a composite model framework through PLS-SEM without compromising predictive accuracy. Astrachan et al. also emphasized that the application of PLS-SEM is particularly effective in managing complex model distributions and numerous indicator variables55. Thus, this study utilizes divergence and convergence validity measures to assess the reliability of the structural model parameters derived from SEM.

Considering the strong recommendation from the PLS literature regarding G*power analysis for determining the optimal sample size56, the sample size estimates were analyzed using G*power 3.1.9.6 software. The sample size of our study was 336, which is notably larger than the minimum requirement of 199 set by the G*power test. This test has an efficacy of 0.8, an alpha value of 0.05, and an effect size of 0.2. In this study, the Pearson’s chi-square test for discrete variables was applied to compare early and late respondents across different variables, including age, educational background, field of study, professional title, joining time, team size, and financial investment. The research demonstrates a lack of significant difference between early and late respondents, effectively eliminating the possibility of non-response bias in this study. Additionally, this study utilizes an ANN model to enhance PLS-SEM, bypassing the latter’s limitation in identifying only corrective and linear relationships57, and ANN supplemented the non-linear relationship in the model of this study, which helps in decision making58. Based on the PLS-SEM analysis, the significant predictors were ranked considering their normalised significance59. The graphical representation of the research methodology is presented in Fig. 2.

Fig. 2
figure 2

Flowchart of research materials and methods.

Full size image

Results

Descriptive analysis

The descriptive analysis of latent structures in Table 2 reveals that the mean scores for IPP, RTI, IM, GP are 3.571, 3.248, 3.579, 3.653, respectively. According to the previous study60, the skewness and kurtosis values of the variables are below the thresholds of ± 3 and ± 10, respectively. The correlation analysis demonstrated that, as depicted in Table 2, only one values among the potential structures exceeded 0.8, and their corresponding coefficients were all below 0.69. This finding suggests that the data does not suffer from severe multicollinearity issues. Additionally, based on the Variance Inflation Factor (VIF), it can be observed that the DV criterion is met, indicating that the model is suitable for further statistical analysis.

Measurement model

This research Utilized four assessments to construct convergent validity, discriminant validity, item-level reliability, and internal consistency reliability. As illustrated in Table 3, the minimal factor loading was 0.625, and the maximal factor reached 0.881. These numbers exceed the suggested threshold of 0.5061. The study indicates the robustness of individual items in this research. If the external loading value lies between 0.40 and 0.50, the researcher can utilize the item without it influencing the retrieved composite reliability (CR) and average variance (AVE) significantly. As per Nunnally and Bernstein’s recommendation, the Cronbach’s alpha and Curreir ratio (CR) of each variable are utilized to determine internal consistency reliability62. It is indicated by Nunnally and Bernstein that a Cronbach’s alpha value should be above the threshold of 0.762. In this study, the Cronbach’s alpha value ranges from 0.769 to 0.974, exceeding the threshold of 0.7, suggesting strong internal consistency63. According to Hair et al., a CR value of more than 0.6 is mandatory61. In the exploratory survey, a CR value between 0.60 and 0.70 is considered acceptable. Values between 0.70 and 0.95 are regarded as good, and those exceeding 0.95 are considered the best. The results in Table 2 reveal that all structures have CR values higher than 0.866, suggesting a model structure with good accuracy and reliability, which meets the internal consistency test64. Convergent validity pertains to the measure of how variable indicators mirror the same underlying structure. The data in Table 3 demonstrates that the minimal AVE value stands at 0.653, whereas the maximum value reaches 0.909. Therefore, this study adheres to the convergent validity standard set by Hair et al., which necessitates an AVE value of no less than 0.50 62.

Table 3 Construct validity and reliability (N = 336).
Full size table

Discriminant validity (DV) is a situation where researchers find it difficult to differentiate between two indicators due to their lack of statistical uniqueness65. Fornell and Larcker introduced two standard indicators for calculating DV through two separate methods66. The first approach involves comparing the square root of AVE with the correlation statistic, while the second approach entails comparing the AVE value with the squared correlation value. In recent years, researchers have devised a novel technique for calculating DV, which shows the shortcomings of the previous metrics. Henseler et al. introduced the heterotrait-monotrait ratio (HTMT) correlation as a novel method for calculating DV67. This study also utilizes the Fornell-Larcker criterion and HTMT criterion. Studies by others suggest that the square root of the AVE for each structure should be greater than the correlation coefficient for each row, as shown in Table 4, to validate the discriminant validity of the structure. According to Henseler et al. it was shown that the theoretical threshold of HTMT is 0.9 for theoretically identical constructs and 0.85 for conceptually different variables67. As depicted in Table 5, all constructions have an HTMT of less than 0.9, except for the HTMT between IM and GP, which is 9.15. Nevertheless, according to Hair et al. (2016) demonstrated is the use of the Variance Inflation Factor (VIF) for assessing multicollinearity, with its value being below 5 being crucial61. The analysis showcases that all VIF values in Table 5 fall beneath this limit, fulfilling the DV criterion. Therefore, the study’s multicollinearity issue is resolved, paving the way for further analyses.

Table 4 Fornell−Larcker criterion.
Full size table
Table 5 HTMT criterion.
Full size table

Table 6 illustrates the predictive relevance of the structure, reflecting the predictive potential of the model’s predictor variables. R2 and Q2, as indicators of predictive power, are employed respectively according to Cohen68; a value of R2 exceeding 0.26 is considered valuable. The R² values for the variables IPP, RTI, IM, in Table 5 are 1, 0.569, 0.980, 0.747, respectively, indicating the variables’ strong predictive abilities. Additionally, Q2’s magnitude signifies the importance of the endogenous component, with a value greater than 0 indicating its relevance in prediction. The study reveals that the Q2 values for the predictive relevance indicators of IPP, RTI, IM, were 0.678, 0.364, 0.659, 0.567, respectively. Additionally, the SRMR values from the PLS-SEM were utilized to evaluate the model’s fit, and the SRMR coefficient for the model in this research was 0.06, which fell short of the 0.1 threshold. Consequently, the model exhibits a satisfactory fit.

Table 6 Predictive relevance of the model.
Full size table

Structural model

After validating the measurement model, the study employed SmartPLS 4.0.9.5 to test the hypotheses associated with the research model. By calculating p-values and t-values, the significance of the proposed hypotheses was assessed. If the t-value surpasses 1.96 or the p-value falls below 0.05, the hypothesis is accepted, otherwise it is rejected. In the current research, we utilized PLS-SEM in conjunction with the bootstrapping method available in SmartPLS to select a bootstrap sample consisting of 5000 observations from the raw data. The purpose of this selection process was to evaluate the significance of the path coefficients, and the findings of the hypothesis testing are depicted in Fig. 3; Table 7.

Fig. 3
figure 3

Structural model.

Full size image
Table 7 Results of hypothesis testing.
Full size table

The results showed that five of the six hypothesized relationships in the model of this study were significant and one hypothesis was not significant. As depicted in Table 7, IPP has a positive impact on NRP team RTI (β = 0.386, t = 5.043、p = 0.000), which supports the hypothesis H1. The path coefficients in Table 7 indicate that a 1% increase in IPP leads to a 0.386% increase in RTI. In addition, IPP has a significant positive effect on for IM (β = 0.030, t = 2.380, p = 0.009), thus supporting H2a. However, the coefficients were found to indicate that IM did not mediate the effect of IPP in the effect of IPP on the RTI of NRPT (β = 0.000, t = 0.065, p = 0.474), and therefore did not support H2b. IPP was found to be a significant predictor of GP (β = 0.099, t = 1.865, p = 0.031), and thus supported H3a. through the observation, GP was found to have a significant mediating effect (β = 0.045, t = 1.819, p = 0.034) in the effect of IPP on the RTI of NRPT, thus supporting H3b. Finally, according to the path coefficients, it was shown that IM and GP had a significant chained mediating effect in the effect of IPP on the RTI of NRPT (β = 0.011, t = 2.192, p = 0.014), thus supporting H4.

ANN analysis

  1. 1.

    Analysis of neural network outputs. The term “ANN” refers to a “massively parallel distributed processor” composed of “simple processing units with neural tendencies to store and utilize experimental knowledge”69. Empirical findings indicate that ANN outperforms conventional regression techniques70. In the current research, ANN analysis was performed with the aid of SPSS26 software, and variables significant according to PLS-SEM were included in the ANN analysis. Consequently, the IPP, RTI, IM and GP variables were considered relevant. Figure 4 displays the ANN model incorporating one output neuron (RTI) and three input neurons (IPP, IM, GP). In this study, a deep ANN with a hidden layer for each output neuron node was utilized58. To improve the prediction accuracy, a sigmoid function was applied to stimulate both output and hidden neurons. The established interval between input and output neurons was set to [0,1]71. To reduce overfitting in the ANN model, this study employed a 90:10 ten-fold cross-validation technique to train and validate the collected data72. The model’s training and testing phases, utilizing the derived data, yielded mean RMSE values of 0.09906 and 0.08867, respectively (as presented in Table 8). The RMSE metrics are utilized to assess the neural network model’s accuracy73. The RMSE is calculated using the formula: RMSE =\(\:\sqrt{\frac{SSE}{N}}\), where SSE denotes the sum-of-square error of the training or test data, and N represents the number of samples for the training or test data. The RMSE values for the test and training data in the ANN model analysis were relatively small, indicating a high predictive accuracy for the ANN model in this study74.

    Table 8 RMSE values.
    Full size table
  2. 2.

    Sensitivity analysis. Subsequently, this study performed a sensitivity analysis to rank the exogenous components according to their normalized relative importance in relation to the endogenous structure, as depicted in Table 9. To determine the normalized normalized importance of each neuron in this research, the relative importance of each neuron was calculated by dividing it by its maximum importance, and this was subsequently reported as a percentage outcome. Among the variables, IPP displays the greatest predictive capability for the RTI of the NRP team. The normalized importance values are 100%, 69.9%, 93.1%, for IPP, GP, IM, respectively.

    Table 9 Sensitivity analysis.
    Full size table
    Fig. 4
    figure 4

    Artificial neural network diagram.

    Full size image

Discussion

This study utilizes Innovation Drive Theory and the AMO framework to empirically examine how IPP fosters RTI within NRPY. The findings from the PLS-SEM and ANN analyses provide robust support for the proposed hypotheses, particularly in confirming H1, where IPP was shown to significantly enhance RTI. According to Innovation Drive Theory, strong institutional mechanisms like IPP are fundamental in sustaining and accelerating innovation. Our empirical results extend this theory by illustrating how IPP not only protects intellectual property but also invigorates the innovation ecosystem, leading to sustained technological advancements. Future research could explore what these additional elements might be, offering a more nuanced understanding of the IPP-RTI dynamic. Particularly, small businesses are more vulnerable to the influence of intellectual property rights protection on their digital innovation compared to large enterprises75.

In examining the relationship between IPP and the IM of NRPY (H2a), our findings align with the “Opportunity” component of the AMO framework, which posits that a supportive environment fosters collaboration and knowledge sharing6. While IM was expected to mediate the IPP-RTI relationship (H2b), our results indicate that this mediation effect was not significant. This suggests that while IPP strengthens the IM, translating this into breakthrough innovation may require additional factors or conditions. The results also show that IPP has a positive impact on GP (H3a). The AMO framework’s “Ability” and “Motivation” components are particularly relevant here, as IPP provides the necessary protection and recognition that motivates team members to fully utilize their abilities, thereby enhancing GP33. The study further demonstrates that GP mediates the relationship between IPP and RTI (H3b), suggesting that when IPP secures intellectual assets, it not only enhances individual contributions but also amplifies the collective capabilities of the team, which are crucial for achieving RTI.

Finally, the chain-mediating effect of IM and GP (H4) adds depth to both Innovation Drive Theory and the AMO framework by showing how these elements interact to magnify the impact of IPP on RTI. Specifically, our findings indicate that the combination of a conducive IM and enhanced GP creates a synergistic effect that significantly boosts the likelihood of breakthrough innovations. This chain mediation underscores the importance of a holistic approach in managing IPP, where fostering both a supportive environment and strong team dynamics are essential for driving significant technological advancements.

Conclusions and implications

Conclusions

The role of IPP in driving organizational innovation has been well-documented. However, this study advances the discourse by examining not only the direct influence of IPP on RTI but also the mediating roles of GP and IM within NRPY. Our findings, grounded in the Innovation Drive Theory and the AMO framework, diverge from earlier works by offering a more comprehensive understanding of how IPP interacts with internal team dynamics to enhance innovation outcomes76. The PLS-SEM and ANN analyses reveal intricate relationships that previous research has overlooked.

The empirical findings from this study confirm that IPP significantly enhances RTI within NRPY, echoing prior research that identified IPP as a key driver of innovation33,75. However, this study extends existing knowledge by demonstrating that the confidence and motivation provided by IPP are not only crucial for protecting intellectual assets but also for facilitating more collaborative and risk-taking behaviors among researchers, leading to transformative advances that benefit society as a whole. Additionally, the findings demonstrate that GP serves as a mediator between IPP and RTI, which is further enhanced by the IM. This chain-mediating effect advances the current understanding by showing how a well-protected IP environment, combined with a conducive IM, enables national research teams to fully harness their collective potential. This insight adds a new dimension to the existing literature, which has often overlooked the interplay between these factors77.

The ANN model results confirm that IPP, GP, and IM are the primary predictors of RTI in NRPY, with IPP being the most influential. This finding aligns with the existing literature but also introduces a new methodological approach by integrating ANN with PLS-SEM to better capture the complex relationships at play78. It also offers a new perspective by ranking these factors according to their predictive strength. The study suggests that a robust IPP system, coupled with effective management of GP and a supportive IM, is crucial for achieving transformative innovation outcomes. This ranking provides actionable insights for policymakers and research managers seeking to prioritize efforts in fostering innovation.

Theoretical implications

In alignment with the global trend towards high-quality economic development driven by science, technology, and innovation, this study contributes to the theoretical discourse by building on the foundational frameworks of Innovation Drive Theory and the AMO Theory. While previous studies have primarily focused on the direct effects of social and organizational behaviors on incremental innovations79, our research shifts the focus to the impact of IPP on RTI. By integrating the AMO Theory, our empirical findings extend the current understanding of how IPP enhance team members’ security, incentives, and opportunities, thereby significantly contributing to the realization of RTI. This shift from incremental to radical innovation marks a critical advancement in the literature.

This study proposes a comprehensive framework for future research on team management within national research projects, specifically addressing the interplay between IPP, IM, and GP in driving RTI. Unlike prior studies that have predominantly focused on single perspectives, this research adopts a multi-dimensional approach, offering a broader understanding of how these factors interact80. While previous literature has often concentrated on incremental innovations or the integration of technological advancements with Industry 4.081, our study shifts the focus to RTI, thus significantly contributing to the enrichment of innovation literature. Furthermore, by examining the chain mediation effects of IM and GP, this research addresses gaps in the existing studies that have largely overlooked these mediating dynamics.

Finally, this study employs a novel methodological approach by combining PLS-SEM with nonlinear analysis through ANN, offering a more nuanced understanding of the causal relationships driving breakthrough technological innovation within teams. While previous research has predominantly relied on linear models82, our study’s integration of nonlinear analysis provides a deeper insight into the complex dynamics at play. The consistency between the results from the nonlinear analysis and the research hypotheses not only validates the theoretical framework but also enhances the robustness of the empirical findings, thereby contributing to a more comprehensive understanding of the antecedents of technological innovation.

Policy and managerial implications

The study underscores the need for detailed and actionable policies to bolster IPP. Governments should focus on crafting comprehensive IP strategies that include, for instance, setting up IP protection task forces to address specific industry challenges, providing tax incentives for firms investing in IP protection, and creating public-private partnerships to enhance IP enforcement. Furthermore, establishing national IP databases and improving transparency in IP management practices can help researchers and organizations navigate the complex landscape of IP laws. These steps will ensure that the benefits of scientific innovations are adequately protected and leveraged for national development. To foster effective collaboration between scientific research clusters and industry, policies should include the establishment of innovation hubs that facilitate direct interactions between researchers and businesses. For example, creating grant programs specifically for joint research projects and providing funding for innovation incubators can bridge the gap between research and commercialization. Additionally, implementing leadership training programs that focus on managing collaborative research efforts and integrating industry insights into scientific projects can enhance the innovation capabilities of research teams. These measures will support the development of RTI and ensure that research outcomes have a tangible impact.

Limitations and future research avenues

The study is not without its shortcomings, and there still exist unexplored avenues that require further investigation. This study presents preliminary findings, which can be used as a reference for further exploration on the relationships between IPP, RTI, IM and GP. However, future research should delve into the aspects related to technological innovation transformation and gain a deeper understanding of various dimensions, including economic environment, political environment, organizational situation, etc. This study primarily focuses on national research projects carried out within universities and research institutes, in contrast to enterprises. The research outcomes from enterprises are more market-relevant and environmentally friendly. Consequently, there is a requirement to enhance the examination of enterprise research project teams. To verify the claims made in this paper, we propose further empirical research on the impact of IPP on RTI. This analysis could be conducted by examining the influence of external factors, such as industry and market demand, regulatory mechanisms, economic conditions, and competitor activities. This study takes into account both internal factors, including quality control system and knowledge-sharing mechanism resource allocation, as well as external influences. Given a cross-sectional research design, we are unable to establish longitudinal connections between the studied structures. Consequently, upcoming research should utilize varied longitudinal study approaches to examine the lasting impact of IPR protection on RTI within NRPT.