Consciousness

How Sentience Emerged From the Ancient Mists of Evolution

The evolutionary path of consciousness from sensing to sentience.

Updated January 29, 2025 | Reviewed by Michelle Quirk

Key points

Neurobiological emergentism proposes that sentience arose via standard physical processes (“weak” emergence).
It emerged along an evolutionary path from early life processes to simple sensing to complex neural systems.
Qualitative aspects of organisms’ experience are inherently subjective—inaccessible to objective explanation.

A Naturalistic Framework for Sentience

The capacity for subjective experience, or sentience, is a defining feature of consciousness. An organism is considered sentient if “there is something it is like to be that organism.”¹ In his 2024 book From Sensing to Sentience,² psychiatrist-neurologist Todd Feinberg focuses on the evolution of sentience—the subjective “feeling” aspects of consciousness—and its personal nature. “Sentient” is derived from the Latin sentire, meaning “to feel.”

The book expands ideas from previous books and articles by Feinberg that were written with his collaborator, the evolutionary biologist Jon Mallatt. Together,³ they developed an evolutionary framework for understanding how early forms of consciousness⁴ emerged as a natural property grounded in life’s basic biological properties but requiring the evolution of more complex neurobiological features for subjective experience to emerge.⁵

In From Sensing to Sentience, Feinberg proposes a new theory he calls neurobiological emergentism that is based upon a biological-neurobiological-evolutionary model that explains both how sentience emerges from complex nervous systems as well as scientifically resolving the “explanatory gap” between the observable physical nervous system and subjective experience.⁶

The theory traces the evolutionary process from the first stages of basic mechanistic sensing in single-celled organisms, which can essentially be thought of as biological machines lacking inner experience,⁷ to primitive animals that have some but not sufficient building blocks for sentience, to those that have a rudimentary form of subjective experience and, thus, simple sentience, to those that are probably fully sentient.

Emergence and Complexity

Feinberg frames sentience as an emergent property, arising from the hierarchical organization of neural systems. Emergent phenomena exhibit novel properties that arise naturally from the interactions of simpler components within a complex system. These properties cannot be fully reduced to the system’s parts; they reflect the organization and complexity of the whole. For Feinberg, sentience is one such emergent property. More specifically, it is a phenomenon of “weak” emergence,⁸ meaning that the emergent properties can, in principle, be explained or derived from the system's underlying components and rules and do not require invoking metaphysical or nonreductive explanations.

Features defining emergent systems:

Complex systems: Emergence arises in systems with many interacting parts, where dynamic processes, not static components, are key to producing new phenomena.
Novel aggregate functions: Emergent properties are new functions that don't exist in individual parts but arise from their collective interactions—like sentience, which emerges from the coordinated activity of neurons in a structured neural system.
Hierarchical organization: Complex systems are hierarchically organized, with higher levels built on interactions at lower levels. These higher levels give rise to novel properties that cannot be reduced to those of their individual components.
Constraints: Constraints, emerging from interactions among lower-level components, maintain stability and coherence in complex systems. By imposing rules or limits on interactions, they shape how components across multiple levels function together, ensuring the system operates as a unified whole.⁹
Reciprocal causality: Interactions within and between levels of a hierarchy are bidirectional: Higher levels guide and ensure global coherence (top-down causality), while lower levels shape and drive higher-level dynamics (bottom-up causality).
Multiple realizability: Emergent phenomena can arise through different configurations of parts. For instance, sentience evolved independently in vertebrates, arthropods, and cephalopods, showing how diverse neural architectures can achieve similar outcomes through shared principles of integration and organization.

The Evolutionary Path to Sentience

In Feinberg’s model, the evolutionary emergence of sentience can be understood through three major stages, each marked by increasing biological and neural complexity:

The first stage, Emergent Stage 1,¹⁰ began approximately 3.5 billion years ago (or possibly earlier) with the appearance of the first recognizable life forms. Single-celled organisms demonstrated basic biological functions such as metabolism and homeostasis, and responsiveness to environmental stimuli via photoreceptors, mechanoreceptors, and chemoreceptors (i.e., mechanistic sensing). These processes enabled organisms to regulate their internal states and interact with the external world. Although these early life forms lacked subjective experience, they exhibited a rudimentary form of “interiority,”¹¹ laying the groundwork for more advanced forms of biological selves.

The second stage, Emergent Stage 2, occurred around 570 million years ago with the evolution of simple nervous systems. Evolving from multicellular animals with diffuse neural nets such as jellyfish, worm-like animals with bilateral symmetry developed more centralized neural circuits (in a head at the front of their bodies), capable of integrating sensory input and coordinating reflexive behaviors. These systems allowed for greater adaptability and more complex interactions with the environment. However, as "presentient" animals, their simplicity limited their capacity to generate unified mental representations or subjective experience.

The third stage, Emergent Stage 3 (ES3), marked the evolutionary emergence of sentience in animals with complex nervous systems. Between 560 and 520 million years ago,¹² vertebrates, arthropods, and cephalopods independently evolved advanced neural architectures capable of integrating sensory and affective information into unified mental states.

Consciousness Essential Reads

How a Mind Emerges From Mindless Things

The Many Lenses of Consciousness

Key innovations of ES3 animals include the following:

Novel Neural Structures and Processes Supporting the Emergence of Sentience

Brains with an increased number of neurons (approximately > 100,000), allowing for higher computational capacity and complexity.
Many differentiated neuronal subtypes, enabling functional specialization and precise neural processing.
Elaborated sensory organs, including image-forming eyes and receptor systems for touch, hearing, and smell, enabling detailed environmental perception (exteroceptive mental images¹³).
More fully developed neural infrastructure for affect and pain, allowing for nonreflexive interoceptive-affective feelings.¹⁴
Expansion of neural hierarchies with extensive reciprocal neural interactions, facilitating coherent integration of diverse inputs across sensory, motor, and affective domains.

Feinberg also lists novel emergent features related to sentience, summarized in Footnote 15.

These innovations collectively allowed ES3 animals (including all vertebrates, some arthropods including certain insects like bees, and cephalopod molluscs such as octopuses) to be capable of subjective experiences and adaptive, goal-oriented behaviors.

Behavioral Markers of Sentience

Feinberg identifies several key behavioral markers, listed in Footnote 16, that can infer the presence of sentience in animals and distinguish it from reflexive responses or simple sensory processes.

Addressing the Explanatory Gap

The so-called "explanatory gap" highlights the challenge of bridging the subjective, first-person nature of experience with the objective, third-person description of the brain. Feinberg proposes that neurobiological emergentism leads to a scientific resolution of what are actually two related “explanatory gaps,” both of which are created by the natural biological emergence of sentience. These gaps are the personal nature of sentience and the qualitative character of experience, the latter being the question of why experiences feel the particular way that they do (also referred to as qualia). However, in place of the “explanatory gaps,” Feinberg proposes that there is actually an experiential gap: The qualitative aspects of an organism's experience are inherently subjective and therefore simply not accessible to objective explanations. This subjective-objective divide is not an obstacle to a physicalist theory of consciousness.¹⁷

Conclusion

In From Sensing to Sentience, Feinberg presents sentience as a “weakly” emergent phenomenon that arose naturally from complex neurobiological systems—which evolved as extensions of the basic biology of living agents. The emergence of sentience was determined by progressive evolutionary increases in essentially four variables: (1) number of neurons, (2) degree of specialized neural functions, (3) number of neurohierarchical levels, and (4) degree of interaction between neurohierarchical levels.¹⁸ This kind of complex system has properties capable of producing higher-level emergent features, by coherently integrating sensory and affective inputs via dynamic reciprocal feedback loops.¹⁹ Sentience is one such emergent feature, a property of the entire system working together—arising from the integration and interaction of very many component processes. Sentience, and more generally consciousness, is the neural system’s unified experience of itself and its relationship to the outside world.

See Footnote 20 for possible ways the theory could be further enhanced and key elements that we need to understand more about.

References

1. Thomas Nagel, "What Is It Like to Be a Bat?" The Philosophical Review 83, no. 4 (1974): 435–450, https://doi.org/10.2307/2183914.

2. Todd E. Feinberg, From Sensing to Sentience: How Feeling Emerges from the Brain (Cambridge, MA: MIT Press, 2024).

3. In this post, I will therefore refer to both authors together at times and will draw from their previous papers (listed in Footnote 5) in addition to Feinberg’s sole-authored 2024 book From Sensing to Sentience.

4. They used the term primary consciousness in their previous work.

5. Todd E. Feinberg and Jon Mallatt, The Ancient Origins of Consciousness: How the Brain Created Experience (Cambridge, MA: MIT Press, 2016); Todd E. Feinberg and Jon M. Mallatt, Consciousness Demystified (Cambridge, MA: MIT Press, 2018); Todd E. Feinberg and Jon M. Mallatt, "Subjectivity 'Demystified': Neurobiology, Evolution, and the Explanatory Gap," Frontiers in Psychology 10 (2019): 1686, https://doi.org/10.3389/fpsyg.2019.01686; Todd E. Feinberg and Jon M. Mallatt, "Phenomenal Consciousness and Emergence: Eliminating the Explanatory Gap," Frontiers in Psychology 11 (2020): 1041, https://doi.org/10.3389/fpsyg.2020.01041.

6. Feinberg and Mallatt referred to their theory as neurobiological naturalism. In From Sensing to Sentience, Feinberg puts more emphasis on the emergence aspect of the theory and therefore calls it neurobiological emergentism.

7. Feinberg rejects single-cell theories of consciousness, such as the cellular basis of consciousness theory.

8. As opposed to a more mysterious and disputed phenomenon referred to as strong emergence, which introduces fundamentally new causal forces that are believed to not be explainable or reducible—even in principle—to physical processes or to the system's underlying components and lower-level rules.

9. In neural systems, for instance, constraints organize sensory maps and pathways, enabling the integration of diverse sensory and affective processes.

10. Feinberg and Mallatt’s previous work refers to levels 1, 2, and 3, while Feinberg’s 2024 book refers to emergent stages 1, 2, and 3.

11. Interiority refers to the way organisms maintain boundaries between themselves (their interior) and the external world, allowing them to regulate their internal states while responding to environmental changes.

12. This was shortly before and during the Cambrian period. The Cambrian was a period of rapid evolutionary diversification about 540-485 mya, thought to have been brought about by an evolutionary “arms race.”

13. Exteroceptive refers to the sensory perception of stimuli originating from the external environment, such as sight, sound, touch, taste, and smell.

14. Nonreflexive refers to responses or feelings that are not merely simple, automatic reflexes. Interoceptive refers to the sensory perception of stimuli originating from within the body, such as hunger, thirst, pain, and internal bodily states including sensations from internal organs. "Affective," in the context of primitive sentient animals, refers to internal "feeling" states that carry positive or negative value, driving behaviors such as approaching beneficial stimuli or avoiding harmful ones. These affective states form the foundation for emotional processes in the most complex animals.

In Feinberg and Mallatt’s theory, sentience is broadly divided into two types: sensory feelings that are generated from the world (exteroceptive sentience) and those that are more internally generated (interoceptive-affective sentience; pain and pleasure, emotions, etc.).

15. Novel emergent features related to sentience:

–Centralized topographical maps, which create exteroceptive "sensory images," providing coherent mental representations of the external environment.

–Centralized positive and negative affects, including pain beyond simple reflexive responses, allowing animals to assign value to stimuli. This facilitates adaptive behaviors, motivation, and decision-making.

–Increasingly nonreflexive (volitional) actions and globally directed behaviors, enabling goal-directed, flexible responses to environmental challenges.

16. Behavioral markers of sentience:

–Motivational trade-offs, where an organism balances competing needs (such as feeding versus avoiding predators), demonstrating the capacity for subjective valuation of priorities. The bumblebee choosing the sucrose-rich yet uncomfortably hot feeder is another example.

–Flexible self-protective behaviors, such as wound-tending or targeted grooming, suggesting an ability to represent specific bodily states shaped by valenced experiences.

–Associative learning, by which animals link stimuli with rewards or punishments, reflecting higher-order cognitive processes tied to subjective experience.

–Other examples of behavioral markers include pain detection and processing of harmful stimuli, as well as a demonstrated preference for pain relief (analgesia preference), which together suggest an ability to respond to harmful stimuli in a unified and internally prioritized way.

These behavioral markers were adapted by Feinberg from the following primary sources: Jonathan Birch et al., Review of the Evidence of Sentience in Cephalopod Molluscs and Decapod Crustaceans (London: LSE Consulting, London School of Economics and Political Science, 2021); Andrew Crump et al., "Animal Sentience Research: Synthesis and Proposals," Animal Sentience 32, no. 31 (2022): Article 1770, https://doi.org/10.51291/2377-7478.1770; Andrew Crump et al., "Sentience in Decapod Crustaceans: A General Framework and Review of the Evidence," Animal Sentience 32, no. 1 (2022): Article 1691, https://doi.org/10.51291/2377-7478.1691.

17. What Feinberg is saying is that the "gap" exists only as a challenge in our explanatory frameworks, not as an actual divide, and is fully reconcilable with a naturalistic scientific explanation of the biological and neurobiological emergence of sentience. It is a matter of epistemology—how we acquire and understand knowledge—rather than a metaphysical or ontological divide between brain processes and subjective experience (i.e., questions of the nature of being and the "building blocks" of reality). Among other things, Feinberg is reframing the gap as an issue of perspective. The gap reflects the dual character of sentience: It is both an objective product of neural processes and inherently subjective/"personal," tied to the unique perspective of each organism (its “interiority”—Footnote 11). Sentience is a physically unmeasurable property that naturally emerges from the organism’s biological and neurobiological complexity.

18. Todd E. Feinberg, From Sensing to Sentience: How Feeling Emerges from the Brain (Cambridge, MA: MIT Press, 2024), Table 4.1.

19. Also linking sensory and affective data with other cognitive processes such as attention, memory, and motivational states, as well as action plans and, of course, motor activity. And all of this continually reshaped by learning from experience, via associative learning processes.

20. Possible ways the theory could be further enhanced, and key elements that we need to understand more about:

–Feinberg and Mallatt’s theory moves the scientific understanding of the evolution of consciousness forward substantially. Perhaps further progress can be achieved in the future by trying to integrate the strongest insights from their theory with those of other theories about the evolution of consciousness, such as Simona Ginsburg and Eva Jablonka’s very influential and intricately formulated UAL (Unlimited Associative Learning) theory. Among other things, UAL focuses in great depth on how organisms learn through interaction with their environment, explaining why consciousness evolved as an adaptive learning mechanism. UAL postulates that learning was in fact the main evolutionary driver of consciousness, and it identifies the capacity for UAL as a transitional marker for the evolution of minimal consciousness.

–There is still much more that needs to be understood about how animals, and at a more complex level, humans, come to develop the full richness of subjective experience, as well as feelings of agency, and (at least in humans), a reflective sense of autobiographical self. Clues to all this probably reside in the recursiveness of thought processes, as has been suggested by many theorists. See for example Douglas R. Hofstadter, I Am a Strange Loop (New York: Basic Books, 2007). Similarly, the neuroscientist Stanislas Dehaene speculates that maybe what makes human cognition unique is "the peculiar way we explicitly formulate our ideas using nested or recursive structures of symbols"—Stanislas Dehaene, Consciousness and the Brain: Deciphering How the Brain Codes Our Thoughts (New York: Viking, 2014), 250.

–Key to all this is also a deeper understanding of What Actually Are Emotions? And the role of memory (see, for example, Joseph LeDoux’s extension of Endel Tulving’s theory of memory to theories of consciousness, in my post Emotions Are Key to Understanding Consciousness, including Footnote 22 of that post).

–And we need to understand more about how mental representations acquire semantic meaning. A key insight in that regard is that, at their most basic level, internal representations have a physical correspondence with what they are mapping—they are correlational maps. And, at more abstract levels, thought is built from analogy-making. Fundamental to all this is the relational nature of information.