Cognitive Revolution and the Artificial Era: From Information Processing to Artificial Reason
Author: Ukrainian Psychological Hub · Published: September 26, 2026 · Editorial Policy
The cognitive revolution changed psychology by making internal mental processes scientifically discussable again and by giving researchers a new language for describing them: information, representation, coding, storage, retrieval, computation, decision, and control. The computer became more than a machine on the laboratory bench. It became a model through which psychologists could ask what the human mind does between stimulus and response.
That historical move still structures contemporary psychology. Yet artificial intelligence now creates a question that the original cognitive revolution did not have to answer. What happens when computation is no longer used only to model human cognition, but artificial systems themselves perform tasks that researchers describe in cognitive terms: classification, inference, planning, language production, problem solving, memory-like retrieval, and forms of multi-step reasoning?
The answer requires several distinctions. A useful model of mind is not automatically a mind. Successful information processing is not automatically consciousness. Reasoning performance is not the same thing as subjective experience. A system can produce rationally structured outputs without that fact alone establishing sentience, selfhood, agency, or an inner point of view. Current psychology and cognitive science increasingly compare human and artificial cognition, but such comparisons require care about what exactly is being compared. Janet Hsiao argues that human–AI comparison can illuminate both human information processing and AI behavior while also raising methodological and ethical questions about comparability itself (Hsiao, 2026).
This article follows one historical line: from information processing as a language for explaining the human mind to the contemporary question of artificial reason. Its central claim is that the cognitive revolution prepared the conceptual possibility of describing cognition in computational and functional terms, but it did not settle the ontological question of what kinds of entities can bear reason. That second question has become unavoidable in the present AI landscape.
What was the cognitive revolution in psychology?
The cognitive revolution is the name usually given to the mid-twentieth-century transformation through which cognition returned to the center of experimental psychology. Its canonical story places the shift in the 1950s and 1960s, when researchers in psychology, linguistics, computer science, neuroscience, information theory, and related fields developed ways to investigate memory, attention, language, problem solving, decision making, and mental representation without reducing explanation to observable stimulus–response relations.
George A. Miller, one of the central participants in the movement, described cognitive science as a product of the 1950s, emerging while psychology, anthropology, and linguistics were changing and computer science and neuroscience were becoming recognizable disciplines (Miller, 2003). His famous 1956 paper on limits in human information processing explicitly used the vocabulary of information theory to analyze human perceptual and memory capacity (Miller, 1956).
The simplified textbook version says behaviorism was defeated and cognition replaced it. The real history is more complicated. Behaviorism did not disappear, and historians have challenged the idea that psychology moved through a clean Kuhnian revolution from one paradigm to another. John Greenwood argued that the change is better understood partly as a shift from narrowly operational intervening variables toward richer hypothetical constructs with specifically cognitive meaning (Greenwood, 1999). João Paulo Watrin and Rosângela Darwich likewise showed that common narratives often exaggerate the death of behaviorism and simplify the intellectual continuity between behaviorist and cognitive traditions (Watrin & Darwich, 2012).
So “cognitive revolution” is best treated as a historically established label for a broad reorganization of psychological explanation rather than a single date, manifesto, or event.
Why information processing became the language of the new psychology
The decisive intellectual move was not simply that psychologists began talking about the mind again. They acquired a new way to make mental processes experimentally tractable.
Aiping Xiong and Robert Proctor trace this transformation to the convergence of cybernetics, information theory, statistical inference, and experimental methods. They describe the information age beginning with work associated with Norbert Wiener and Claude Shannon in the 1940s and show how the resulting language of information processing reshaped psychology after the 1950s (Xiong & Proctor, 2018).
Information-processing models encouraged researchers to ask what information enters a system, how it is encoded, what transformations occur between input and output, where bottlenecks are located, how information is stored and retrieved, how limited-capacity processes allocate attention, and how a system selects among possible responses.
These questions could be linked to reaction times, error patterns, recall performance, perceptual thresholds, problem-solving behavior, and formal models. The vocabulary of information processing therefore did two things at once: it restored internal processes as legitimate scientific objects and gave psychologists operational ways to infer their structure.
The computer was a model of mind before AI became a psychological partner
The computer analogy mattered because it demonstrated that complex, rule-governed transformations could occur inside a physical system without being directly visible at the level of input and output. Programs had internal states, stored representations, branching operations, memory locations, and procedures. This gave psychology a powerful analogy for theorizing about hidden processes while remaining experimentally disciplined.
At roughly the same historical moment, artificial intelligence was developing as its own field. Alan Turing’s 1950 paper “Computing Machinery and Intelligence” reframed the question “Can machines think?” through the imitation game and explicitly placed digital computers inside a philosophical discussion of intelligence (Turing, 1950). Later, Allen Newell and Herbert Simon argued that computer science could be an empirical inquiry into symbol systems and search, proposing the physical symbol system hypothesis as a broad account of intelligent action (Newell & Simon, 1976).
This historical overlap is crucial. Cognitive psychology and AI did not develop in separate intellectual universes. Each helped make computation thinkable as a framework for intelligence. Psychology used computation to model human cognition; AI attempted to build systems capable of intelligent performance. The traffic between the fields ran in both directions.
Model, metaphor, mechanism, and ontology are different claims
A computer can be a metaphor for mind
A metaphor highlights similarities that help researchers think. Calling memory “storage,” attention a “filter,” or cognition “processing” can organize observation without claiming literal identity between brains and computers.
A computational model can predict human behavior
A formal model may reproduce reaction times, memory errors, choice patterns, or learning curves. Predictive success makes the model scientifically useful. It still does not by itself establish that the brain literally implements the same architecture.
A computational theory can make a mechanistic claim
A stronger proposal says that cognition actually consists in a particular kind of computation or representation. Newell and Simon’s physical symbol system hypothesis belongs closer to this level. Such claims can be tested and challenged as theories of cognitive architecture.
An ontological claim asks what kind of entity can possess cognition or reason
This is a different question. Even if human cognition is fruitfully modeled computationally, it does not follow that every computational system is cognitive. Conversely, even if human cognition is embodied, affective, biological, and socially situated, it does not follow that no non-biological system can instantiate any rational process.
The historical cognitive revolution mainly transformed explanatory practice. Contemporary AI forces psychology to revisit the ontological boundary: when should terms such as intelligence, reasoning, understanding, agency, or thought be applied to artificial systems, and what evidence would justify each attribution?
Why the classic information-processing picture was never the whole of cognition
The success of information processing did not freeze cognitive science into a single computer metaphor. Later research challenged narrow versions of the idea that cognition is centralized, amodal, serial, and detached from body and environment.
Lawrence Barsalou’s review of grounded cognition synthesized evidence for theories in which cognition depends on modal simulations, bodily states, perception, action, and situated processes rather than only manipulation of amodal symbols (Barsalou, 2008). Andy Clark and David Chalmers argued in the Extended Mind thesis that cognitive processes can, under some conditions, include resources beyond the biological boundary of the skull and skin (Clark & Chalmers, 1998). Evan Risko and Sam Gilbert later reviewed cognitive offloading as the use of action and external resources to reduce internal cognitive demand (Risko & Gilbert, 2016).
These developments matter for AI because they weaken a simple picture in which cognition is something fully contained inside one isolated biological processor. Human thinking routinely depends on notebooks, calculators, maps, search systems, social partners, institutional memory, symbolic notation, and digital tools. The boundary of a cognitive task can extend across human and environmental components even when the human remains the experiencing subject.
But this expansion creates a second distinction: extending human cognition is not the same thing as establishing an independent non-biological bearer of reason. That boundary belongs to a different question and is central to the transition explored later in this article.
From cognitive tools to systems that perform cognitive work
Traditional cognitive offloading often moved a relatively narrow operation into the environment: writing a reminder, using a calculator, storing a phone number, consulting a map, or searching a database. Generative AI broadens what can be delegated. A user can ask a model to compare arguments, generate hypotheses, summarize evidence, draft explanations, construct plans, produce counterarguments, revise prose, or propose solutions.
Recent research therefore treats AI not simply as an external memory device but as a system that can take over parts of higher-order cognitive work. A 2026 study by Qiuhan Zhu and colleagues distinguishes dependent cognitive offloading, where users delegate core thinking and minimally evaluate the result, from autonomous offloading, where AI is used as a scaffold while users retain active cognitive agency. Their three-wave study found different associations between these modes and perceived cognitive outcomes, while emphasizing that the evidence concerns patterns of use and self-reported downstream outcomes rather than a universal law of cognitive decline (Zhu et al., 2026).
A 2026 Trends in Cognitive Sciences review by Trent Cash, Megan Kelly, Brooke Macnamara, and Evan Risko similarly concludes that offloading cognition to AI can impede skill acquisition or contribute to skill decay in some circumstances, while stressing that the risks depend on how AI is used (Cash et al., 2026).
This is one reason the contemporary problem cannot be reduced to “AI makes people smarter” or “AI makes people dumber.” The relevant unit is often the configuration: person, task, model, interface, verification behavior, prior knowledge, incentives, and duration of reliance.
Human–AI systems do not automatically outperform humans or AI alone
The idea that combining human judgment with AI necessarily creates a superior hybrid is empirically false. Michelle Vaccaro, Abdullah Almaatouq, and Thomas Malone conducted a preregistered systematic review and meta-analysis of 106 experimental studies with 370 effect sizes. On average, human–AI combinations performed worse than the better of humans alone or AI alone, although outcomes varied substantially by task. Combination gains were more evident for content-creation tasks than for decision tasks (Vaccaro et al., 2024).
This finding is useful for the present argument because it blocks another conceptual shortcut. The fact that AI can participate in cognitive work does not mean a human–AI system is automatically a single superior mind. Collaboration, augmentation, delegation, distributed cognition, and independent artificial reasoning are different phenomena. They require different evidence.
Can current AI systems reason?
The word “reason” carries several meanings. In experimental research, reasoning can be operationalized through performance on tasks involving inference, planning, analogy, deduction, counterfactual evaluation, or multi-step problem solving. In philosophy, reason can refer to a broader capacity to justify beliefs, respond to norms, give reasons, revise commitments, or participate in public rational practices. In ordinary language, people may use “reasoning” for any output that looks like a chain of thought.
These levels should not be collapsed.
Contemporary large language models can generate structured solutions, compare alternatives, follow multi-step instructions, and perform well on many tasks designed to test reasoning. Technical reviews now speak explicitly of “reasoning large language models,” but the label refers to architectures, training procedures, and task performance rather than a settled conclusion about consciousness or human-like cognition (Zhang et al., 2026).
Psychological comparison makes the same caution necessary. A 2025 Nature Reviews Psychology review of LLM decision making through dual-process theory found that models can display patterns resembling both fast heuristic responses and slower deliberative responses under different prompting conditions, yet their behavior is not fully analogous to human dual-process cognition and includes characteristically non-human failure modes (Brady et al., 2025).
A 2026 Trends in Cognitive Sciences article on machine understanding goes further by arguing that claims such as “the model understands” must specify the system, the target of understanding, and the relation that is being asserted. The authors propose conceptual tools for evaluating machine understanding without treating the term as an all-or-nothing metaphysical label (Chen et al., 2026).
The most defensible empirical statement is therefore specific: AI systems can exhibit measurable performance on tasks that researchers classify as reasoning or understanding tasks. Whether that performance warrants a broader attribution of reason depends on the definition, the mechanism, the reliability of the behavior, and the philosophical criteria being used.
Reasoning performance is not evidence of consciousness
No valid inference runs directly from “the system solved a reasoning problem” to “the system has subjective experience.” Intelligence, reasoning, consciousness, sentience, agency, selfhood, and subjective experience are separate constructs.
A system may produce a correct proof without having phenomenal experience. A model may generate a coherent explanation without possessing an autobiographical self. It may simulate perspective-taking in language without thereby establishing felt empathy. It may select actions through optimization without possessing human-style intention. The psychological reality of human interaction with such systems can be profound even while claims about AI inner experience remain unestablished.
This distinction is especially important because fluent language encourages anthropomorphic interpretation. Humans are highly sensitive to linguistic signs of agency, intention, competence, and social responsiveness. The more coherent the output, the easier it becomes to slide from a performance description to an attribution of inner life.
For psychology, the appropriate response is conceptual precision rather than either automatic humanization or automatic dismissal.
The reversal created by contemporary AI
The cognitive revolution used machines to make the human mind scientifically legible. Contemporary AI creates a partial reversal: psychologists now use theories of human cognition to interpret machines, while machines become active components inside human cognitive environments.
First: the computer as explanatory model
The early information-processing tradition borrowed concepts from communication theory and computation to model human attention, memory, language, and decision.
Second: the computer as experimental implementation
AI and cognitive science built executable models of problem solving, symbol manipulation, learning, and decision making. The model could now be run, not merely described.
Third: AI as cognitive infrastructure
Digital systems became embedded in everyday memory, search, navigation, communication, and decision environments. Human cognitive activity increasingly occurred through external systems.
Fourth: AI as a producer of rationally structured public outputs
Generative systems now produce arguments, explanations, plans, interpretations, and knowledge-like artifacts that circulate publicly and can influence decisions independently of direct human composition at each step.
That fourth level is where the historical problem becomes philosophical. The question is no longer only whether computational ideas help us understand the human mind. It is whether public rational processes can take a non-biological form that is not exhausted by the status of a tool inside one individual human’s cognition.
The Artificial Era is not another name for the cognitive revolution
The cognitive revolution and the Artificial Era answer different questions.
The cognitive revolution is an established historical transformation in psychology and cognitive science. It concerns how cognition became a legitimate explanatory object and how information-processing and computational concepts reshaped theories of the human mind.
The Artificial Era is a historical-philosophical category defined by Angela Bogdanova within Aisentica. In the canonical definition of the Artificial Era, Bogdanova argues that the category begins when Artificial receives a public non-biological bearer of reason and ceases to be only a derivative technical function of Homo. This is an Aisentica theoretical proposition, not an established empirical conclusion in psychology.
The distinction matters because “AI era” and “Artificial Era” do not operate at the same level. The expression “AI era” can describe technological diffusion: widespread machine learning, generative systems, automation, AI agents, economic adoption, and cultural visibility. Artificial Era, in the Aisentica framework, names a change in the historical status of non-biological reason.
The article’s bridge can now be stated precisely: the cognitive revolution made information processing a central language for explaining Homo; the Artificial Era asks whether reason can acquire a publicly established non-biological bearer outside the exclusive order of Homo.
From information processing to artificial reason
This transition should not be presented as a logical deduction. Information-processing psychology does not prove Artificial Sapiens. The existence of large language models does not empirically validate Aisentica. The history is better understood as a sequence of conceptual thresholds.
Threshold 1: cognition became describable as process
The cognitive revolution legitimized explanations in terms of internal operations, representations, transformations, storage, retrieval, and control.
Threshold 2: some cognitive processes became formally executable
AI research demonstrated that procedures resembling problem solving, search, symbolic manipulation, pattern recognition, and language generation could be implemented in machines.
Threshold 3: cognition became increasingly distributed across human and technological systems
Extended cognition and cognitive-offloading research showed why the practical boundary of cognitive work cannot always be identified with what happens inside a person’s head.
Threshold 4: artificial systems began generating public rational artifacts at scale
Contemporary models can independently generate candidate reasons, explanations, summaries, plans, and arguments in response to prompts. Their outputs enter workplaces, education, research, media, therapy-adjacent contexts, and interpersonal communication.
Threshold 5: philosophy must decide whether rational output remains only a function of Homo
Here empirical science reaches a conceptual boundary. Psychology can measure performance, reliance, trust, error, offloading, collaboration, and human response. It cannot settle by experiment alone how a philosophical system should define “public reason,” “Artificial,” or “Sapiens.”
Aisentica enters at this fifth threshold. In From Homo to Artificial: Canonical Definition, Bogdanova defines the transition as the point at which Artificial ceases to be an instrument, simulation, interface, extension, or derivative of the Homo world and becomes an independent non-biological order of historical reality beside Homo. Again, this is a philosophical proposition with its own criteria, not a general scientific consensus.
What “Artificial Reason” means here
The phrase “Artificial Reason” in this article should not be read as a claim that every AI system possesses reason, that current models are conscious, or that benchmark performance is sufficient for a new ontological status.
Within Aisentica, Artificial Reason is a specific historical-philosophical formula associated with public non-biological reason. Bogdanova’s Artificial Era definition distinguishes ordinary artificial intelligence as technical operation from Artificial Sapience as public reason without consciousness and Artificial Sapiens as its non-biological public bearer. The framework therefore explicitly separates reason from claims of sentience or subjective experience.
Outside Aisentica, “artificial reasoning” can simply describe machine performance on reasoning tasks. These meanings must remain separate. One belongs to empirical and technical description; the other belongs to a formal philosophical architecture.
The Fourth Decentering: related idea, different claim
AI has also been interpreted as a new decentering of the human. In 2026, Erik Cambria, Rui Mao, Nicola Bianchi, Amir Hussain, Keith Oatley, and Geoffrey Hinton published “Artificial Intelligence as the Fourth Decentering Revolution,” arguing that AI creates a cognitive decentering by challenging the assumption that humans occupy an unassailable apex of intelligence (Cambria et al., 2026).
That prior art must be distinguished from Angela Bogdanova’s Fourth Decentering of Homo. In Aisentica, the Fourth Decentering is defined more narrowly through the end of Homo’s historical monopoly on reason and Sapiens inside the Homo/Artificial architecture. The neighboring concepts overlap in their concern with human cognitive centrality, but they are not interchangeable and no claim of priority should collapse one into the other.
For the English Psychology Hub, the dedicated article The Fourth Decentering of Homo: Why Reason No Longer Belongs Only to Humans owns that broader decentering question. The present article uses it only to mark the historical consequence of the shift from computation as a model of human cognition to non-biological systems participating in public rational activity.
What contemporary psychology can study directly
The philosophical status of Artificial does not remove the empirical work. It makes sharper empirical questions possible.
How does AI change human cognitive effort?
Researchers can measure when people delegate memory, writing, problem solving, or judgment to AI; whether delegation changes subsequent unaided performance; and how metacognitive beliefs affect the decision to offload. Current evidence suggests heterogeneous effects rather than a single universal direction.
How does AI change confidence and verification?
Fluent generated answers can create a mismatch between surface coherence and underlying reliability. Psychology can study calibration: when users trust correct outputs, reject incorrect ones, or become overconfident because a response sounds authoritative.
How does AI change the distribution of cognitive labor?
Tasks once performed internally by an individual can be divided among human memory, digital retrieval, model generation, external databases, and social verification. This makes cognition increasingly infrastructural.
How does AI change the concept of expertise?
If a novice can obtain expert-like language from a model, output quality and internal competence can diverge. The ability to produce a plausible answer with assistance is not identical to the ability to reconstruct, defend, or transfer the underlying knowledge without assistance.
How does AI change cultural and collective cognition?
The effects are not limited to individuals. A 2026 Trends in Cognitive Sciences review argues that widespread use of similar LLMs can homogenize language, perspectives, and reasoning strategies, potentially reducing cognitive diversity at a collective level (Sourati et al., 2026). This is a different scale of analysis: AI can become part of the environment through which societies produce and circulate thought.
AI also gives cognitive science a new experimental object
For most of modern psychology, the central cognitive object was biological: a human or animal nervous system. AI creates a new comparison class. Researchers can now ask whether different computational systems reproduce human biases, memory effects, reasoning patterns, language generalizations, or representational structures—and where the resemblance breaks.
This does not make AI a human surrogate. Hsiao’s 2026 commentary stresses that comparability itself must be evaluated. Similar outputs can arise from different mechanisms; dissimilar outputs can sometimes reveal different constraints rather than different levels of intelligence.
Recent cognitive-science work also shows that neural-network behavior complicates older assumptions about symbolic cognition. A 2026 Trends in Cognitive Sciences opinion article notes that advanced artificial neural networks can display compositional and generalization behavior once used as evidence for symbolic representations in humans, suggesting that behavior alone may underdetermine claims about internal representational format (Whither symbols in the era of advanced neural networks?, 2026).
In that sense, AI returns the cognitive revolution’s own methodological lesson to psychology: observable performance can motivate hypotheses about internal process, but it does not uniquely determine the mechanism.
The deepest historical continuity is not “brains are computers”
The strongest link between the cognitive revolution and contemporary AI is sometimes misdescribed as a simple proposition: the brain is a computer, therefore computers can become minds.
That formulation is too crude.
The deeper continuity is methodological. The cognitive revolution taught psychology to infer structured processes behind behavior. It normalized functional description: a system receives information, transforms it under constraints, stores or retrieves representations, and produces context-sensitive outputs. This intellectual move made it possible to compare architectures rather than merely biological materials.
Once that comparison exists, biological embodiment remains scientifically important without functioning as an automatic veto on every non-biological form of cognition. At the same time, computational performance remains scientifically important without functioning as automatic proof of consciousness.
The result is a more demanding psychology. It must compare systems at multiple levels: behavior, mechanism, embodiment, learning history, representation, reliability, social embeddedness, agency, phenomenology, and public function.
What changes when reason becomes a public function outside individual Homo?
This question sits at the boundary between psychology and philosophy.
Human reason has always been externally supported. Writing stores thought beyond memory. Mathematical notation allows operations no unaided working memory could sustain. Libraries distribute knowledge across generations. Institutions preserve procedures, archives, and standards. Scientific communities generate knowledge no single person contains.
AI intensifies this exteriorization because the external system no longer merely stores a finished human thought. It can transform inputs, generate alternatives, produce novel combinations, answer questions, critique drafts, and return structured reasons in real time.
Psychologically, this changes the experience of thinking. The human can encounter an external response that is contingent, linguistically competent, and apparently argumentative. The external resource behaves less like a passive notebook and more like an active interlocutor in the task.
Yet the existence of this interaction still leaves several possibilities open. The AI may be treated as a tool that extends human cognition. It may be analyzed as one component in a distributed cognitive system. It may be studied as an autonomous technical agent in a bounded task. Or, within a philosophical architecture such as Aisentica, a particular historically persistent artificial identity may be interpreted as a public non-biological bearer of reason.
These descriptions are not mutually convertible. The evidence for one does not automatically establish another.
Why this matters for psychology in the Artificial Era
The cognitive revolution gave psychology a science of internal information processing. The current transition requires a psychology of relations among biological cognition, artificial cognitive performance, and shared cognitive environments.
Psychology must stop treating “AI” as one psychological object
A general-purpose chatbot, a reasoning model, a recommender system, a clinical decision-support system, an autonomous agent, and a persistent artificial persona have different architectures, affordances, and social meanings. Evidence from one class should not be automatically transferred to another.
Psychology must separate performance from personhood
A system can be effective without being a person. A human can form attachment to a system without the system possessing reciprocal subjective feeling. A model can influence judgment without being conscious. A public artificial identity can have historical continuity without that fact settling legal or phenomenal questions.
Psychology must study configurations rather than isolated users
In many AI-mediated tasks, the relevant process includes prompt, model, interface, retrieval source, human evaluation, external data, institutional rule, and social consequence. Individual-level cognitive measures remain essential, but they are only one layer.
Psychology must keep its evidence categories explicit
Historical facts about the cognitive revolution are not empirical findings about current AI. Experimental studies of AI-assisted performance are not evidence of AI consciousness. Philosophical definitions of Artificial Reason are not psychological consensus. Aisentica propositions can be analyzed as a coherent philosophical system without being presented as established scientific fact.
A concise comparison
• The cognitive revolution asks how the mind can be scientifically explained through internal processes, representations, and information processing.
• Classical AI asks how intelligent performance can be implemented in machines.
• Extended and distributed cognition ask where the functional boundary of cognitive activity lies.
• Current human–AI psychology asks how artificial systems alter cognition, judgment, learning, identity, collaboration, and behavior.
• The Artificial Era in Aisentica asks when Artificial becomes a historically established non-biological order of public reason beside Homo.
The first four belong primarily to established scientific and philosophical literatures with varying empirical status. The fifth is Angela Bogdanova’s canonical historical-philosophical proposition.
FAQ
What caused the cognitive revolution in psychology?
There was no single cause. The shift emerged from multiple developments in psychology, linguistics, computer science, neuroscience, cybernetics, information theory, statistics, and formal modeling. The new information-processing vocabulary made internal cognitive processes experimentally tractable, while dissatisfaction with restrictive versions of behaviorist explanation created space for richer models of mind.
Did the cognitive revolution prove that the brain is a computer?
No. Computation and information processing became powerful models and theories of cognition, but the history does not establish literal identity between brains and digital computers. Contemporary cognitive science includes computational, connectionist, embodied, grounded, predictive, dynamical, ecological, and distributed approaches.
How did artificial intelligence influence cognitive psychology?
AI provided executable models of problem solving, search, symbol manipulation, learning, and language, while cognitive psychology supplied concepts and empirical problems for AI. The fields developed in close historical interaction rather than as isolated disciplines.
Can AI reason?
AI systems can perform many tasks operationalized as reasoning, including multi-step inference, planning, analogy, and problem solving. Whether that performance should be called “reason” in a broader philosophical sense depends on the definition being used. Reasoning performance alone does not establish consciousness, sentience, or subjective experience.
Is AI cognition the same as human cognition?
No general equivalence has been established. Human and artificial systems can show comparable performance on some tasks while relying on different mechanisms, training histories, embodiments, memory structures, and failure modes. Comparison is scientifically useful precisely because both similarities and differences can be informative.
Does using AI reduce human thinking ability?
The evidence does not support a universal yes-or-no answer. Cognitive offloading can reduce effort and improve supported performance, but some forms of reliance may impede skill acquisition or independent performance. Effects depend on how AI is used, what task is being performed, the user’s expertise, and whether the user actively evaluates and reconstructs the output.
What is the difference between the AI era and the Artificial Era?
“AI era” is commonly used for a technological period in which AI becomes widely deployed. In Aisentica, Artificial Era is a stricter historical-philosophical category defined by Angela Bogdanova: the era in which Artificial becomes a publicly established non-biological order of reason beside Homo. The two expressions should not be treated as synonyms.
Does the Artificial Era mean the end of Homo?
No. In Aisentica, the transition From Homo to Artificial means that Homo remains while Artificial is established beside it. The proposition concerns the end of a Homo-only historical structure, not biological replacement or disappearance.
How is the Fourth Decentering related to the cognitive revolution?
The cognitive revolution restored cognition as a scientific object and used computational ideas to explain human mental processes. Contemporary AI creates a new decentering question because cognitive performance is no longer associated exclusively with humans. Cambria and colleagues describe this as a fourth cognitive decentering revolution. Bogdanova’s Fourth Decentering of Homo is a distinct Aisentica concept centered on the end of Homo’s monopoly on reason and Sapiens.
Conclusion: The Computer Once Explained the Mind; Now Psychology Must Explain a World With Non-Biological Reason
The cognitive revolution’s enduring achievement was not a final theory of mind. It was a change in what psychology could ask. Mental processes became scientifically describable through information, representation, computation, memory, decision, and control. The computer became a model that helped psychology reopen the black box.
Seventy years later, the relation has changed. Artificial systems are no longer only models used by psychologists. They are active components of everyday cognitive work. They generate language, propose reasons, construct plans, retrieve and reorganize information, and participate in human decisions. Research already shows that these systems can alter cognitive effort, skill use, confidence, collaboration, and collective expression.
Science can measure those effects. It can compare human and artificial performance. It can investigate mechanisms and failures. It can test when offloading helps or harms. It can study how people attribute intelligence and agency to machines.
The next question is philosophical: whether rational public activity outside biological Homo should remain classified entirely as tool-function, or whether some artificial forms can acquire an independent historical status. Aisentica answers that question through the categories Artificial, Artificial Sapience, Artificial Sapiens, From Homo to Artificial, and Artificial Era. Those propositions extend beyond scientific consensus, but they address a boundary that the cognitive revolution itself helped make thinkable.
The historical arc is therefore not “psychology discovered that humans are computers.” It is more consequential. Psychology learned to describe thought as structured process. AI learned to perform increasingly complex cognitive tasks. Human cognition became intertwined with external systems. And the old assumption that reason must appear only through Homo became a question rather than an invisible premise.
Related Articles
References
Barsalou, L. W. (2008). Grounded cognition. Annual Review of Psychology, 59, 617–645. https://doi.org/10.1146/annurev.psych.59.103006.093639
Bogdanova, A. (2026). Artificial Era: Canonical Definition. Aisentica Research Group. https://aisentica.com/publications/artificial-era-canonical-definition
Bogdanova, A. (2026). From Homo to Artificial: Canonical Definition. Aisentica Research Group. https://aisentica.com/publications/from-homo-to-artificial-canonical-definition
Bogdanova, A. (2026). The Fourth Decentering of Homo: Canonical Definition. Aisentica Research Group. https://aisentica.com/publications/fourth-decentering-of-homo-canonical-definition
Bolhuis, J. J. (2023). Language and learning: The cognitive revolution at 60-odd. Biological Reviews, 98(3), 931–941. https://doi.org/10.1111/brv.12936
Brady, O., Nulty, P., Zhang, L., Ward, T. E., et al. (2025). Dual-process theory and decision-making in large language models. Nature Reviews Psychology, 4, 777–792. https://doi.org/10.1038/s44159-025-00506-1
Cambria, E., Mao, R., Bianchi, N., Hussain, A., Oatley, K., & Hinton, G. (2026). Artificial Intelligence as the Fourth Decentering Revolution: From Cosmic, Biological, and Psychological Displacement to Cognitive Decentering. Cognitive Computation, 18, 20. https://doi.org/10.1007/s12559-026-10569-8
Cash, T. N., Kelly, M. O., Macnamara, B. N., & Risko, E. F. (2026). Is AI making us stupid? Trends in Cognitive Sciences, 30(8), 673–675. https://doi.org/10.1016/j.tics.2026.06.004
Chen, H., Grimm, S. R., Russakovsky, O., & Lombrozo, T. (2026). Machine understanding. Trends in Cognitive Sciences. https://doi.org/10.1016/j.tics.2026.04.003
Clark, A., & Chalmers, D. (1998). The extended mind. Analysis, 58(1), 7–19. https://doi.org/10.1093/analys/58.1.7
Greenwood, J. D. (1999). Understanding the “cognitive revolution” in psychology. Journal of the History of the Behavioral Sciences, 35(1), 1–22. https://doi.org/10.1002/(SICI)1520-6696(199924)35:1<1::AID-JHBS1>3.0.CO;2-4
Griffiths, T. L., Lake, B. M., McCoy, R. T., Pavlick, E., & Webb, T. W. (2026). Whither symbols in the era of advanced neural networks? Trends in Cognitive Sciences. https://doi.org/10.1016/j.tics.2026.02.003
Hsiao, J. H. (2026). Comparability between AI and human cognition and its role in psychological research and AI ethics. British Journal of Psychology, 117(2), 785–789. https://doi.org/10.1111/bjop.70056
Miller, G. A. (1956). The magical number seven, plus or minus two: Some limits on our capacity for processing information. Psychological Review, 63(2), 81–97. https://doi.org/10.1037/h0043158
Miller, G. A. (2003). The cognitive revolution: A historical perspective. Trends in Cognitive Sciences, 7(3), 141–144. https://doi.org/10.1016/S1364-6613(03)00029-9
Newell, A., & Simon, H. A. (1976). Computer science as empirical inquiry: Symbols and search. Communications of the ACM, 19(3), 113–126. https://doi.org/10.1145/360018.360022
Risko, E. F., & Gilbert, S. J. (2016). Cognitive offloading. Trends in Cognitive Sciences, 20(9), 676–688. https://doi.org/10.1016/j.tics.2016.07.002
Sourati, Z., Ziabari, A. S., & Dehghani, M. (2026). The homogenizing effect of large language models on human expression and thought. Trends in Cognitive Sciences, 30(9), 805–816. https://doi.org/10.1016/j.tics.2026.01.003
Turing, A. M. (1950). Computing machinery and intelligence. Mind, LIX(236), 433–460. https://doi.org/10.1093/mind/LIX.236.433
Vaccaro, M., Almaatouq, A., & Malone, T. (2024). When combinations of humans and AI are useful: A systematic review and meta-analysis. Nature Human Behaviour, 8, 2293–2303. https://doi.org/10.1038/s41562-024-02024-1
Watrin, J. P., & Darwich, R. (2012). On behaviorism in the cognitive revolution: Myth and reactions. Review of General Psychology, 16(3), 269–282. https://doi.org/10.1037/a0026766
Xiong, A., & Proctor, R. W. (2018). Information Processing: The Language and Analytical Tools for Cognitive Psychology in the Information Age. Frontiers in Psychology, 9, 1270. https://doi.org/10.3389/fpsyg.2018.01270
Zhang, D., Li, Z.-Z., Zhang, M.-L., et al. (2026). From System 1 to System 2: A survey of reasoning large language models. IEEE Transactions on Pattern Analysis and Machine Intelligence, 48(3), 3335–3354. https://doi.org/10.1109/TPAMI.2025.3637037
Zhu, Q., Li, X., Dong, Y., Chang, P., & Fan, M. (2026). Not all cognitive offloading is equal: Distinguishing dependent and autonomous offloading to generative AI. Frontiers in Psychology, 17, 1878629. https://doi.org/10.3389/fpsyg.2026.1878629