Introduction
What does it mean to feel, to sense, to reason? For millennia, we’ve considered the depth and nuance of these experiences to be uniquely human. We now stand at a new frontier—one where that assumption is being profoundly and creatively challenged. This frontier is Artificial Psychology (AP; Picard, 1997; Russell & Norvig, 2021), a theoretical and practical transformative advancement at the intersection of artificial intelligence and psychological science (Farahani et al., 2023).
It dares to move beyond traditional analysis to envision intelligent systems capable of considering complex mind data. To do this, the field unfolds across two beautiful, complementary dimensions: Approximate Psychology and Intricate Psychology (Farahani et al., 2023; 2024; Farahani, Watson, & Bazden, 2025; Kosko, 1997).
First, Approximate Psychology looks at our human side and acknowledges the inherent fluidity of our thought and emotion. Think about it—our feelings aren’t simple on/off switches. Our perceptions and feelings fluctuate with context and uncertainty. Our best decisions often arise from a gut feeling—an intuition—rather than a strict calculation. This “fuzziness,” our remarkable tolerance for ambiguity and paradox, is something conventional computational models have always struggled with. By combining fuzzy set theory (Zadeh, 1965), probabilistic reasoning (Pearl, 2009), and adaptive machine learning, approximate psychology allows an AI to represent mental states with graceful uncertainty (Farahani et al., 2025; Kovač et al., 2024). This approach softens mechanical rigidity, evolving our understanding of psychology from fixed equations toward dynamic, living patterns.
Second, Intricate Psychology takes on the role of the architect, constructing the deep architecture of the mind. It computationally replicates the internal choreography of perception, memory, language, reasoning, and emotion that defines how we think (Farahani, Watson, & Bazden, 2025; Friston et al., 2017).
This is precisely why Artificial Psychology is so necessary. As a highly multidisciplinary field, it tries to solve the problems that occur when psychologists need robust analytic frameworks. Both classical statistics and machine learning represent complementary approaches to the extraction of knowledge from data (Kovač et al., 2024; Shalev Shwartz & Ben David, 2014).
Yet both paradigms share a fundamental epistemic limitation: they operate only on observable data. Unobserved variables, latent structures, and hidden dynamics remain outside their reach (Pearl & Mackenzie, 2018). Consequently, both systems model a projection of reality, not reality itself. This is where promoting and improving artificial intelligence helps us find a holistic model of mental models (Farahani, Kovač, Fardi, & Watson, 2025; Friston, 2022).
“Overlaps, evolution, and blurring are the norm.” By fusing theoretical fuzzy inference, deep learning algorithms, and probabilistic models, we can finally offer authentic, flexible models capable of sophisticated prediction—and more importantly—of engaging real world psychological complexity (Farahani et al., 2023; Kovač et al., 2025; Russell & Norvig, 2021).
This evolution is not merely methodological but epistemological (Farahani et al., 2025; Friston et al., 2017). It redefines what it means to know a mind—shifting from static measurement to dynamic, ongoing understanding. We already see applications such as adaptive learning systems that respond to a student’s mood and socially assistive robots that modulate affective tone (Cowie & Cornelius, 2022; Farahani et al., 2023; Picard, 1997).
Human psychology is not impersonal; it reflects diverse cultural contexts and arises from deeply personal data. Artificial psychology, therefore, must remain both scientifically rigorous and ethically vigilant (Farahani et al., 2023; Floridi & Cowls, 2021). Its progress depends on sustained dialogue among engineers, clinicians, philosophers, and artists—a dialogue ensuring simulation of thought while honoring personhood.
Recognizing this shift, our university has established a Master of Science (M.S.) in Artificial Psychology to train experts fluent in both psychological science and computational intellect (Farahani et al., 2023). Despite its later renaming to “Psychology and Artificial Intelligence,” the foundational vision continues to shape its theoretical core.
Hojjatollah Farahani, PhD
Peter Watson, Phd
Forough Esrafilian, PhD
Abbasali Allahyari, PhD
Nataša Kovač, PhD
Nina Hubig, PhD
Timea Bezdan, PhD
Mariјa Blagoјević, PhD
Sara Saljoughi, PhD 
