Psychological Capital as the Catalyst: A PROCESS Moderated Mediation Model of AI Risk Perception and Ethical Intervention in ECE

https://doi.org/10.5281/zenodo.18031988

Abstract

Background: The proliferation of algorithmic tools in Early Childhood Education (ECE) raises urgent ethical concerns regarding algorithmic bias (AB) and its potential to exacerbate inequities. ECE teachers are the frontline defense against these biases, yet the transition from recognizing bias to executing corrective Ethical Intervention Behavior (EIB) often fails, highlighting an acute awareness-action gap. This study investigates the complex, conditional psychological pathway governing ECE teachers' ethical responses.

Methodology: Employing a quantitative, cross-sectional design, this research utilized the Moderated Mediation Model (Hayes PROCESS Model 8). Data were collected from N=304 ECE teachers in Lahore and Islamabad. The model tested the indirect effect of Perceived Algorithmic Bias (AB) on EIB via AI Ethical Literacy (AEL), with Psychological Capital (PsyCap) hypothesized as a moderator influencing the AEL  EIB path. All variables were measured using reliable scales  0.88).

Key Findings: The analysis yielded conclusive evidence for a conditional process. First, the direct effect of AB on EIB was statistically insignificant (B=0.05, p=0.327). Second, the crucial interaction term between AEL and PsyCap was highly significant (B_interaction =0.18, p=0.005). This confirmed that the efficacy of ethical knowledge is moderated by internal resources. Critically, the conditional indirect effect was found to be significant only under conditions of moderate and high PsyCap (95% CI excluded zero), confirming that PsyCap acts as the enabling factor that allows cognitive competency (AEL) to successfully translate into proactive behavioral intervention.

Conclusion and Implication: We conclude that ethical action among ECE teachers is a conditional psychological process. For teachers to intervene effectively against algorithmic bias, AI Ethical Literacy must be augmented by Psychological Capital (efficacy, resilience, optimism, and hope). The findings mandate a policy shift towards holistic teacher development, prioritizing both technical ethical training and the cultivation of psychological resilience to foster genuine ethical agency in the AI-mediated classroom.

Keywords: Algorithmic Bias, AI Ethical Literacy, Psychological Capital, Ethical Intervention Behavior, Moderated Mediation, ECE.

Background: The proliferation of algorithmic tools in Early Childhood Education (ECE) raises urgent ethical concerns regarding algorithmic bias (AB) and its potential to exacerbate inequities. ECE teachers are the frontline defense against these biases, yet the transition from recognizing bias to executing corrective Ethical Intervention Behavior (EIB) often fails, highlighting an acute awareness-action gap. This study investigates the complex, conditional psychological pathway governing ECE teachers' ethical responses.

Methodology: Employing a quantitative, cross-sectional design, this research utilized the Moderated Mediation Model (Hayes PROCESS Model 8). Data were collected from N=304 ECE teachers in Lahore and Islamabad. The model tested the indirect effect of Perceived Algorithmic Bias (AB) on EIB via AI Ethical Literacy (AEL), with Psychological Capital (PsyCap) hypothesized as a moderator influencing the AEL  EIB path. All variables were measured using reliable scales  0.88).

Key Findings: The analysis yielded conclusive evidence for a conditional process. First, the direct effect of AB on EIB was statistically insignificant (B=0.05, p=0.327). Second, the crucial interaction term between AEL and PsyCap was highly significant (B_interaction =0.18, p=0.005). This confirmed that the efficacy of ethical knowledge is moderated by internal resources. Critically, the conditional indirect effect was found to be significant only under conditions of moderate and high PsyCap (95% CI excluded zero), confirming that PsyCap acts as the enabling factor that allows cognitive competency (AEL) to successfully translate into proactive behavioral intervention.

Conclusion and Implication: We conclude that ethical action among ECE teachers is a conditional psychological process. For teachers to intervene effectively against algorithmic bias, AI Ethical Literacy must be augmented by Psychological Capital (efficacy, resilience, optimism, and hope). The findings mandate a policy shift towards holistic teacher development, prioritizing both technical ethical training and the cultivation of psychological resilience to foster genuine ethical agency in the AI-mediated classroom.