Towards Neurocognitive-Inspired Intelligence: From AI's Structural Mimicry to Human-Like Functional Cognition

TL;DR

The paper argues that current AI systems are narrow, data-hungry, and opaque, lacking generalization, memory, and real-time adaptability. It proposes Neurocognitive-Inspired Intelligence (NII), a biologically grounded, modular framework that integrates perception, memory, attention, reasoning, and action within closed perception-action loops and embodied contexts. The authors detail a seven-module architecture and outline concrete implementation strategies, learning regimes, and evaluation metrics, illustrated with proof-of-concept experiments in multimodal perception, resilient robotics, and cognitive-health monitoring. They also discuss scalability, neuromorphic hardware considerations, and a path toward real-world deployment in domains like robotics, healthcare, education, and safety. Overall, NII aims to deliver human-like generality, data-efficient learning, and transparent reasoning through integrated cognitive modules and neurosymbolic inference, enabling AI systems that sense, plan, and learn in dynamic environments.

Abstract

Artificial intelligence has advanced significantly through deep learning, reinforcement learning, and large language and vision models. However, these systems often remain task specific, struggle to adapt to changing conditions, and cannot generalize in ways similar to human cognition. Additionally, they mainly focus on mimicking brain structures, which often leads to black-box models with limited transparency and adaptability. Inspired by the structure and function of biological cognition, this paper introduces the concept of "Neurocognitive-Inspired Intelligence (NII)," a hybrid approach that combines neuroscience, cognitive science, computer vision, and AI to develop more general, adaptive, and robust intelligent systems capable of rapid learning, learning from less data, and leveraging prior experience. These systems aim to emulate the human brain's ability to flexibly learn, reason, remember, perceive, and act in real-world settings with minimal supervision. We review the limitations of current AI methods, define core principles of neurocognitive-inspired intelligence, and propose a modular, biologically inspired architecture that emphasizes integration, embodiment, and adaptability. We also discuss potential implementation strategies and outline various real-world applications, from robotics to education and healthcare. Importantly, this paper offers a hybrid roadmap for future research, laying the groundwork for building AI systems that more closely resemble human cognition.

Figures (14)

• Figure 1: Illustrating neurocognitive inspired intelligence framework, providing a high-level overview of the core components of neurocognitive intelligence and their functional roles.
• Figure 2: Neurocognitive-Inspired Intelligence (NII) framework with a dedicated Learning Module. Inputs (e.g., sensory data, language, images) are encoded by the Perception Unit and prioritized by the Attention Mechanism. Salient information is stored/retrieved by the Memory Module and refined by the Learning Module, which updates representations and priors used by the Reasoning Engine. The Adaptation Layer modulates strategies based on prediction error, confidence, and context, and the Action/Output Unit executes decisions. Feedback from actions is routed back to perception and into the cognitive core (memory--learning--reasoning), enabling continual improvement, error correction, and context-aware control.
• Figure 3: Perception unit. A hierarchical module that transforms raw sensory inputs (visual, auditory, tactile, proprioceptive) into mid- to high-level abstractions via biologically inspired encoders, active perception, and symbolic grounding.
• Figure 4: Attention mechanism. A bidirectional, context-sensitive attention controller, highlighting how top-down goals and bottom-up salience modulate cognitive resource allocation.
• Figure 5: Memory module. A dual-memory system inspired by biological memory structures, supporting working memory, episodic recall, semantic abstraction, and lifelong learning through consolidation and synaptic plasticity.