philosophy of mind

In the philosophy of mind, functionalism posits that every mental state—such as beliefs, desires, or pain—is exclusively defined by its functional role. This role encompasses its causal interactions with other mental states, sensory inputs, and observable behaviors. Functionalism emerged primarily as a counter-proposal to both the identity theory of mind and behaviorism.

Functionalism operates as a theoretical framework situated between physical implementation and behavioral manifestations. Consequently, it diverges from antecedent theories such as Cartesian dualism, which posits distinct mental and physical substances, and Skinnerian behaviorism and physicalism, which assert the existence of only physical substances. Functionalism's focus is exclusively on the brain's effective functions, conceptualized through its organizational structure or its "software programs."

Given that a mental state is characterized by its functional role, it is considered multiply realizable; that is, it can manifest across diverse systems, potentially including computers, provided these systems execute the requisite functions. A computer program achieves these functions through computations on inputs to generate outputs, facilitated by its electronic substrate, whereas a brain accomplishes them via its biological operations and responses to stimuli.

Multiple Realizability

The concept of multiple realizability constitutes a significant component of certain arguments supporting functionalism. Within standard functionalist frameworks, a mental state is equated with a specific functional role. This can be analogized to a valve, which can be constructed from various materials like plastic or metal, yet consistently performs its essential function of regulating fluid or gas flow. Analogously, functionalists contend that a mental state can be elucidated without reference to the underlying physical substrate, such as the brain, that instantiates it; instead, only its higher-level functions require consideration. Since a mental state is not confined to a singular physical medium, it possesses the capacity for multiple instantiations, theoretically extending to non-biological systems like computers. Consequently, a machine based on silicon could potentially exhibit a mental life comparable to that of a human, contingent upon its structure realizing the appropriate functional roles.

Although the majority of functionalist theories endorse the multiple realizability of mental states, certain functionalist frameworks, such as Functional Specification Theories (FSTs), explicitly reject this premise. FSTs were prominently advanced by David Lewis and David Malet Armstrong. Under the FST perspective, mental states are defined as the specific "realizers" of a functional role, rather than the functional role itself. For instance, the mental state of belief is identified with the particular brain or neurological process that actualizes the corresponding belief function. Therefore, in contrast to standard functionalist approaches, often termed Functional State Identity Theories, FSTs preclude the multiple realizability of mental states, asserting that their realization by brain states is a fundamental characteristic. This viewpoint is frequently motivated by the conviction that if an alien species possessed a cognitive system materially distinct from humans (e.g., silicon-based) yet performed identical functions to human mental states (e.g., vocalizing "Ouch!" when subjected to sharp stimuli), their mental states would be considered similar but not identical to ours. For some, this constitutes a drawback of FSTs. Notably, Hilary Putnam's argument for his own functionalist model leveraged the intuition that such extraterrestrial beings would indeed share the same mental states as humans, thereby positioning the multiple realizability inherent in standard functionalism as a superior theory of mind.

Types

Machine-State Functionalism

"Functionalism," as a broad philosophical stance, encompasses numerous distinct interpretations. Hilary Putnam introduced the initial formulation of a functionalist theory of mind during the 1960s. This particular framework, now designated as machine-state functionalism or simply machine functionalism, drew inspiration from parallels identified by Putnam and others between the human mind and the theoretical "machines" or universal computers capable of executing any algorithm, conceptualized by Alan Turing as Turing machines. By the mid-1970s, Putnam himself started to challenge this very position, with his "Twin Earth" thought experiment marking the inception of his dissent from machine-state functionalism.

Conceptually, a Turing machine represents an abstract construct grounded in a mathematical model, rather than a tangible physical entity. Characteristically, a Turing Machine incorporates a horizontal tape segmented into rectangular cells, extending from left to right. This tape possesses infinite length, and each individual cell is capable of holding a symbol, with the specific set of symbols varying across different "machines." A read-write head is integral to the machine, enabling it to scan cells and traverse both left and right. The machine's operations are dictated by the symbol present in the currently scanned cell and a predefined table of transition rules, which effectively constitute its programming. Due to the tape's infinite nature, a conventional Turing Machine theoretically possesses unlimited time to compute any specific function or multiple functions. In the subsequent illustration, each cell is either empty (B) or contains the symbol §45§, serving as the machine's inputs. The potential outputs include:

• Halt: Cease all operations.
• R: Advance one cell to the right.
• L: Advance one cell to the left.
• B: Erase the content of the current cell.
• 1: Erase the content of the current cell and inscribe a '1'.

A straightforward illustration involves a Turing machine designed to write the sequence '111' after processing three blank squares, subsequently halting as defined by the ensuing machine table:

The provided table specifies that if the machine is in state one and encounters a blank square (B), it will print a §23§ and persist in state one. Should it be in state one and read a §45§, it will shift one square to the right and transition to state two. If the machine is in state two and reads a B, it will print a §89§ and maintain state two. Conversely, if it is in state two and reads a §1011§, it will move one square to the right and enter state three. Furthermore, if it is in state three and reads a B, it will print a §1415§ and continue in state three. Lastly, if the machine is in state three and reads a §1617§, it will remain in state three.

The crucial aspect to apprehend here pertains to the nature of the states within the Turing machine. Each state is exclusively definable by its relationships to other states, as well as to its inputs and outputs. For instance, state one is characterized solely by the machine's action of writing a §45§ and remaining in that state upon reading a B, or by its action of moving one square to the right and transitioning to a different state upon reading a §67§. This constitutes the functional definition of state one, representing its causal contribution within the entire system. The specific mechanisms by which it achieves its functions and its underlying material composition are entirely immaterial.

The preceding observation is fundamental for comprehending machine-state functionalism. Given that Turing machines are not inherently physical systems, "anything capable of undergoing a temporal succession of states can be considered a Turing machine." Consequently, since biological organisms inherently “go through a succession of states in time,” it follows that such organisms could also be deemed equivalent to Turing machines.

Under the tenets of machine-state functionalism, the intrinsic character of a mental state is analogous to that of the Turing machine states previously delineated. If the rational operations and computational capabilities of these machines can be demonstrated as comparable to those of human beings, then it logically follows that Turing machine behavior closely approximates human behavior. Hence, the specific machine or mental state is not attributable to a particular physical-chemical composition, but rather to the programming rules that generate the observed effects. Stated differently, any rational preference arises from adherence to rules, independent of the agent's specific material constitution.

Psychofunctionalism

A distinct form of functionalism emerges from the repudiation of behaviorist psychological theories, advocating instead for empirical cognitive models of mental processes. This perspective, primarily linked to Jerry Fodor and Zenon Pylyshyn, is designated as psycho-functionalism.

Psycho-functionalism posits that psychology constitutes an irreducibly complex scientific discipline. Consequently, the terminology employed to characterize mental entities and properties within robust psychological theories cannot be reduced to simple behavioral dispositions; moreover, such a reduction would be neither desirable nor significant, even if feasible. Proponents of psychofunctionalism contend that psychology, akin to the biological sciences, utilizes inherently teleological or purposive explanations. For instance, the heart's function is to circulate blood, while the kidney's role involves filtration and the maintenance of chemical equilibrium; these functional definitions are crucial for scientific explanation and classification. Although these mechanisms may manifest in diverse physical forms, their significance lies solely in their contribution to the overarching biological theory. Similarly, the function of mental states, including beliefs and desires, is delineated by the functional or causal role assigned to them within the most robust scientific psychological frameworks. Should a mental state posited by folk psychology (e.g., hysteria) lack a fundamental explanatory role in cognitive psychology, its existence may be negated. Conversely, if theoretical cognitive psychology necessitates states for explaining human behavior that are not encompassed by ordinary folk psychological language, then such entities or states are deemed to exist.

Analytic functionalism

A third variant of functionalism addresses the semantics of theoretical terms broadly. This perspective, primarily linked to David Lewis, is frequently termed analytic functionalism or conceptual functionalism. The core tenet of analytic functionalism asserts that theoretical terms derive their implicit definitions from the theories within which they are formulated, rather than from the intrinsic properties of their constituent phonemes. For ordinary language terms like "belief," "desire," or "hunger," their meanings are understood to originate from common-sense "folk psychological" theories. However, these conceptualizations are deemed inadequate to withstand the stringent demands of materialistic theories concerning reality and causality. Consequently, these terms undergo conceptual analyses, typically structured as follows:

Mental state M is defined as the state that is antecedently conceived by P and subsequently causes Q.

For instance, the experience of pain is caused by an event such as sitting on a tack, and subsequently causes overt expressions like loud cries, alongside higher-order mental states such as anger and resentment directed towards the individual responsible for the tack. Such functional definitions, articulated through causal roles, are posited as analytic and a priori truths pertaining to submental states and the (often hypothetical) propositional attitudes they characterize. Consequently, adherents of this view are identified as analytic or conceptual functionalists. The fundamental distinction between analytic functionalism and psychofunctionalism lies in the latter's emphasis on laboratory observation and experimentation. This empirical approach is crucial for ascertaining the authenticity of mental state terms and concepts, and for identifying functional equivalences as genuinely contingent and a posteriori identities. Conversely, analytic functionalism asserts that these identities are necessary and thus not amenable to empirical scientific inquiry.

Homuncular functionalism

Homuncular functionalism was developed largely by Daniel Dennett and has been advocated by William Lycan. It emerged in response to the conceptual challenges posed by Ned Block's China Brain (also known as the Chinese nation) and John Searle's Chinese room thought experiments, which targeted conventional functionalist theories. To address the conceptual complexities stemming from the hypothetical scenario of a nation of individuals, each functioning as a single neuron within a collective system to generate the functional mental states of an individual mind, many functionalists accepted the implication. They contended that such a 'Chinese nation' would indeed exhibit all qualitative and intentional properties of a mind, thereby constituting a systemic or collective mind endowed with propositional attitudes and other mental attributes. Regardless of the merit of this hypothesis, an immediate objection arose concerning its implication of an unacceptable form of mind-mind supervenience. According to Block's formulation, the systemic mind, purportedly emerging at a higher level, would necessarily supervene on the individual minds of each member within the Chinese nation. This proposition, however, appears to significantly challenge, if not directly contradict, the core principle of the supervenience thesis: that alterations in the mental domain must correspond to changes in the underlying physical substratum. This concept becomes evident when designating the set of higher-level mental facts as M1 and the set of lower-level mental facts as M2. In this framework, both M1 and M2 supervene on physical facts. However, a transition from M1 to M2, for instance, could theoretically transpire without any corresponding alteration in these physical facts.

Given the apparent acceptance of mind-mind supervenience within functionalist discourse, some theorists proposed that the resolution to this dilemma lay in positing a hierarchical series of mental levels, analogous to homunculi. These levels would progressively diminish in sophistication regarding functional organization and physical composition, ultimately reaching the level of physico-mechanical neurons or neuronal groups. According to this perspective, the homunculi at each successive level possess genuine mental properties, yet they exhibit increasing simplicity and reduced intelligence further down the hierarchy.

Mechanistic Functionalism

Mechanistic functionalism, independently developed and advocated by Gualtiero Piccinini and Carl Gillett, enhances prior functionalist descriptions of mental states by asserting that all psychological explanations must be framed mechanistically. Consequently, rather than mental states being explained solely through their functional relationships with other mental states, functions are considered only one component of a mental state's explanation, with structures constituting the other essential part.

A mechanistic explanation entails dissecting a system, such as a mental system, into its constituent physical components, their respective activities or functions, and their integrated organizational relationships. Under this framework, the mind persists as a functional system, yet it is interpreted through mechanistic principles. This perspective retains its functionalist character because functional relations remain crucial for mental states. However, it is mechanistic because these functional relations consistently manifest as concrete structures, even if these structures are conceptualized at a particular level of abstraction. Functions are individuated and elucidated either by their contributions to the specified system or through teleological considerations. When understood teleologically, functions can be further characterized as either etiological or non-etiological.

Mechanistic functionalism shifts the focus of functionalism from the traditional autonomy of psychology relative to neuroscience, instead promoting the integration of these two disciplines. By offering a practical framework for synthesizing conventional psychological models with neurological data, mechanistic functionalism can be seen as harmonizing the functionalist theory of mind with neurological explanations of brain function. This reconciliation stems from mechanistic explanations of function, which endeavor to elucidate the physical realization of functional states (mental states) via neurological mechanisms.

Physicalism

Significant ambiguity surrounds the purported relationship between the overarching theory of functionalism and physicalism. Functionalism has frequently been asserted to invalidate physicalism tout court, meaning without further qualification. Conversely, a majority of philosophers of mind who endorse functionalism also identify as physicalists; notably, some, like David Lewis, have even declared themselves strict reductionist physicalists.

Ned Block characterizes functionalism primarily as a broadly metaphysical thesis rather than a narrowly ontological one. Consequently, functionalism's central concern is not with the nature of existence but rather with the defining characteristics of specific mental states, such as pain, as distinct types of states. Prior efforts to address the mind-body problem sought to resolve it by answering both questions: dualism posited two substances and defined mental states by their immateriality; behaviorism asserted a single substance, characterizing mental states as behavioral dispositions; and physicalism maintained the existence of one substance, defining mental states as physical states (e.g., "pain = C-fiber firings").

From this perspective, type physicalism appears incompatible with functionalism, given that type physicalism asserts mental states (e.g., pain) are characterized by their physical nature, whereas functionalism defines pain by its functional/causal role and its connection to behaviors like exclaiming "ouch." Nevertheless, any less stringent form of physicalism, which merely posits the ontological claim that all existing entities are composed of physical matter, remains entirely compatible with functionalism. Furthermore, most functionalists who also identify as physicalists stipulate that the properties referenced in functional definitions must themselves be physical properties. Consequently, they are physicalists, even if the broader functionalist thesis does not inherently necessitate this commitment.

David Lewis distinguishes between the concepts of "having pain," which functions as a rigid designator applicable to the same entities across all possible worlds, and "pain" itself, a non-rigid designator. For Lewis, "pain" denotes a definite description, such as "the state with causal role x." In humans, the referent of this description is a specific type of brain state, subject to scientific identification. For silicon-based life forms, the referent would differ, and for angels, it would be an immaterial, non-physical state. Thus, Lewis posits that local type-physical reductions are feasible and align with conceptual functionalism. There appears to be an unresolved ambiguity between types and tokens within functionalist analysis that warrants clarification.

Criticism

A 2020 PhilPapers survey indicated functionalism as the most favored theory, with 33% of participants accepting or leaning towards it, followed by dualism at 22%, and identity theory at 13%. Despite this prevalence, functionalism presents counter-intuitive implications, frequently challenged through the use of thought experiments.

China Brain

Ned Block critiques the functionalist concept of multiple realizability, which posits that hardware implementation is inconsequential, with only the functional level holding significance. The "China brain" or "Chinese nation" thought experiment postulates that if the entire population of China were systematically organized to function as a brain, with each individual acting as a neuron, then—provided the people perform the appropriate functional roles and maintain the correct causal relationships between inputs and outputs—the system would constitute a genuine mind, complete with mental states and consciousness. (The significant disparity in operational speed among individual units is typically overlooked.) Ned Block asserts the implausibility of this scenario, suggesting a fundamental flaw in the functionalist thesis if it permits such a system to be deemed a legitimate mind.

Some functionalists posit that China could possess qualia, yet its immense scale renders the concept of a conscious Chinese nation unimaginable. This perspective suggests that human understanding of consciousness might be limited by our inherent theory of mind, preventing comprehension of a national consciousness. Consequently, if functionalism holds true, qualia must either manifest in any system executing the appropriate functions, irrespective of its physical composition, or they are entirely illusory.

The Chinese Room Argument

John Searle's Chinese Room argument directly challenges the assertion that thought can be conceptualized as a collection of functions. This thought experiment posits that a purely functional system can simulate intelligent behavior without genuine interpretation or comprehension. Searle illustrates this by describing an English-speaking individual confined to a room containing Chinese symbols and an English rulebook for manipulating them. External Chinese speakers provide symbols, and the individual follows the rulebook to return specific symbols. Searle contends that attributing Chinese language knowledge to the English speaker, solely based on these syntactic operations, would be illogical. The experiment aims to demonstrate that systems relying exclusively on syntactic processes—defined by algorithmic inputs and outputs—cannot achieve semantics (meaning) or intentionality (aboutness). Therefore, Searle refutes the notion that thought is equivalent to adhering to a set of syntactic rules.

A frequent counter-argument to Searle's thought experiment proposes that a higher-level mental activity exists beyond the individual, necessitating consideration of the entire system. This perspective implies that the system itself comprehends Chinese, even if the person within the room does not. The individual is often likened to a central processing unit (CPU) within a computational framework. Searle, in rebuttal, suggested that the person could internalize all rules and symbol relationships. He maintained that even if the individual memorized and executed these operations mentally, they would still be manipulating symbols devoid of inherent meaning. Conversely, some critics argue that this symbol-processing subsystem of the brain could be conceptualized as a distinct, virtual mind capable of understanding Chinese.

Functionalists further contend that, theoretically, a digital hardware system could emulate every neuron in the brain of a Chinese speaker. Such a brain emulation, they argue, would possess identical mental processes and consequently comprehend Chinese.

The Inverted Spectrum Argument

A significant critique of functionalism is the inverted spectrum, or inverted qualia, scenario, notably advanced by Ned Block as an objection. This thought experiment posits an individual, Jane, born with a condition causing her to perceive colors as inverted compared to typical perception; for instance, she sees violet as yellow and orange as blue. Consider, for example, observing an orange with Jane. While you perceive the fruit as orange, Jane perceives it as blue. Nevertheless, both you and Jane would verbally identify the fruit's color as "orange." Crucially, all behavioral and functional interactions with colors would remain identical. Jane, for instance, would correctly respond to traffic signals, despite her altered color perception. The argument concludes that because two functionally identical individuals can possess distinct mental states—specifically differing in their qualitative or phenomenological aspects—functionalism lacks the explanatory power to account for individual variations in qualia.

David Chalmers posits that all "functionally isomorphic" systems, defined by their identical "fine-grained functional organization" or information processing, will possess qualitatively indistinguishable conscious experiences. He designates this concept as the principle of organizational invariance. This principle suggests, for instance, that a silicon chip functionally isomorphic to a human brain would experience the same perception of the color red when exposed to identical sensory inputs. To illustrate this, Chalmers introduced the "dancing qualia" thought experiment. This argument employs a reductio ad absurdum approach, commencing with the premise that two such systems could exhibit distinct qualia under identical circumstances. The experiment envisions a mechanism that alternates between a segment of the brain responsible for perceiving red and a functionally isomorphic system, such as a silicon chip, that elicits the perception of blue. Given that both systems execute the same function within the brain, the subject would remain unaware of any transition during the switch. Chalmers contends that such a scenario would be highly improbable if the qualia were genuinely shifting between red and blue, thereby leading to a contradiction. Consequently, he concludes that the dancing qualia phenomenon is practically unfeasible, and an equivalent digital system would not only manifest qualia but would also possess conscious experiences qualitatively identical to those of its biological counterpart, such as perceiving the same color. Furthermore, Chalmers proposed a related thought experiment, termed "fading qualia," which asserts the impossibility of qualia diminishing when individual biological neurons are progressively substituted with functional equivalents.

A pertinent criticism leveled against the inverted spectrum argument is its foundational assumption that mental states, particularly those distinguished by their qualitative or phenomenological attributes, can exist independently of the brain's functional relationships. This premise, therefore, presumes the conclusion regarding functional mental states, effectively precluding the very possibility of functionalism without providing any autonomous substantiation. Functionalism, in contrast, posits that mental states arise from the functional interconnections within the brain. This identical methodological flaw—characterized by an underlying antithetical assumption rather than a reasoned argument—is also discernible in both the Chinese room and the Chinese nation arguments.

Twin Earth

Hilary Putnam's Twin Earth thought experiment constitutes a principal argument employed against functionalism, despite its initial formulation as a critique of semantic internalism. This straightforward thought experiment unfolds as follows: Envision a Twin Earth that mirrors Earth in every aspect save one—water on Twin Earth possesses a chemical structure, for instance XYZ, distinct from Earth's H₂O. Crucially, however, XYZ on Twin Earth is still designated "water" and manifests all the macroscopic properties exhibited by H₂O on Earth, such as being a transparent, potable liquid found in lakes and rivers. Given that these two worlds are identical in all respects except for the fundamental chemical composition of water, you and your Twin Earth counterpart perceive precisely the same phenomena, interact with the same individuals, hold identical occupations, and behave in an identical manner. Consequently, sharing identical sensory inputs, behavioral outputs, and interrelations among mental states renders you functional duplicates. For instance, both you and your duplicate believe that water is wet. Nevertheless, the semantic content of your mental state concerning the belief that water is wet diverges from your duplicate's, as your belief pertains to H§4_{5§O, whereas your duplicate's refers to XYZ. The argument thus concludes that since two individuals can be functionally identical yet possess disparate mental states, functionalism is inadequate for comprehensively explaining all mental states.}

Initial proponents of functionalism often addressed this argument by attempting to establish a clear differentiation between internal and external content. For instance, the internal content of propositional attitudes would encompass only those elements unrelated to the external world and possessing the requisite functional/causal attributes that facilitate connections with other internal mental states. However, as no definitive foundation or justification for such a distinction in mental content has been articulated, this concept has largely been superseded by externalist causal theories of mental contents, also known as informational semantics. Jerry Fodor's "asymmetric causal theory" of mental content exemplifies this perspective, which involves adapting functionalism to incorporate a broad understanding of inputs and outputs, thereby including external objects that cause mental representations.

The twin earth argument relies on the premise that encountering an imitation water would elicit a distinct mental state compared to experiencing natural water. Nevertheless, given that individuals would not perceive any difference between the two types of water, this premise is probably incorrect. Furthermore, this fundamental assumption directly contradicts functionalism; consequently, the twin earth argument does not present a valid critique, as its core premise inherently rejects functionalism (which posits that the two waters would not generate different mental states because their functional relationships would persist unchanged).

Meaning Holism

A frequent critique of functionalism asserts that it necessitates a radical interpretation of semantic holism. Block and Fodor termed this the damn/darn problem. The distinction between exclaiming "damn" or "darn" after striking one's finger with a hammer can hold mental significance. However, if, according to functionalism, these outputs are connected to numerous (if not all) internal mental states, then two individuals experiencing identical pain but producing different verbal reactions would possess minimal (or no) commonality across their mental states. This conclusion is counterintuitive, as it appears evident that two people share a substantial aspect of their mental state of pain if both injure their finger with a hammer, irrespective of the specific word they utter in distress.

An alternative approach to this issue involves embracing a moderate, or molecularist, form of holism. Yet, even if this proves effective for pain, it encounters challenges when applied to beliefs and meaning, specifically in delineating relevant from non-relevant contents. This distinction is often difficult to establish without resorting to an analytic–synthetic distinction, which many scholars aim to circumvent.

Triviality Arguments

Ned Block contends that for functionalism to circumvent the chauvinism inherent in type-physicalism, it must avoid being excessively permissive in "ascribing mental properties to things that do not in fact have them." He illustrates this by suggesting that Bolivia's economy could be structured so that its economic states, inputs, and outputs are isomorphic to those of a person, given a peculiar mapping between mental and economic variables.

Hilary Putnam, John Searle, and other scholars have advanced arguments positing that functionalism is trivial, implying that the internal structures it seeks to analyze are ubiquitous. This ubiquity suggests that functionalism either devolves into behaviorism or into complete triviality, thereby leading to a form of panpsychism. These arguments commonly assume that physics generates a sequence of unique states and that functionalist realization occurs whenever a mapping exists from a proposed set of mental states to the physical states of a system. Since the states of any physical system are invariably at least marginally unique, such a mapping will always be present, thus rendering any system a mind. Functionalist formulations that impose absolute requirements on interactions with external objects (i.e., objects not functionally defined within the functional account) are reduced to behaviorism rather than absolute triviality, as input-output behavior remains a prerequisite.

Peter Godfrey-Smith contends that these formulations can be rendered trivial if they incorporate a seemingly innocuous additional premise. This premise posits that the inclusion of a transducer layer—an input-output system—to an object should not alter its capacity for mental states. The transducer layer's function is limited to generating behavior based on a straightforward mapping, such as a lookup table, which translates inputs into system actions and system states into outputs. Nevertheless, given that the system occupies unique states at every moment and for every potential input, such a mapping is invariably present, thereby allowing for a transducer layer capable of producing any desired physical behavior.

Godfrey-Smith proposes that these issues might be resolved through the application of causality, suggesting a continuum between objects possessing mental states and those lacking them, rather than a rigid dichotomy. Moreover, imposing constraints on these mappings appears to necessitate either an examination of external behavior, akin to behaviorism, or an analysis of the internal structure of the realization, as seen in identity theory. While multiple realizability may not be entirely compromised, the functionalist assertion regarding the autonomy of high-level functional descriptions becomes debatable.

References

References

Armstrong, D.M. (1968). A Materialistic Theory of the Mind. London: RKP.

• Armstrong, D.M. (1968). A Materialistic Theory of the Mind. London: RKP.
• Baron-Cohen, S.; Leslie, A.; Frith, U. (1985). "Does the Autistic Child Have a "Theory of Mind"?". Cognition. 21 (1): 37–46. doi:10.1016/0010-0277(85)90022-8. PMID 2934210. S2CID 14955234.Readings in Philosophy of Psychology. Cambridge, MA: Harvard University Press.
• Block, Ned. (1980b). "Troubles With Functionalism", in Block (1980a).
• Block, Ned. (1994). Qualia. In S. Guttenplan (ed.), A Companion to Philosophy of Mind. Oxford: Blackwell.
• Block, Ned (1996). "What is functionalism?" (PDF). A revised version of the entry on functionalism in The Encyclopedia of Philosophy Supplement, Macmillan.Philosophical Review 81.
• Chalmers, David. (1996). The Conscious Mind. Oxford: Oxford University Press.
• DeLancey, C. (2002). "Passionate Engines – What Emotions Reveal about the Mind and Artificial Intelligence." Oxford: Oxford University Press.
• Dennett, D. (1990). Quining Qualia. In W. Lycan (ed.), Mind and Cognition. Oxford: Blackwells.
• Levin, Janet. (2004). "Functionalism", The Stanford Encyclopedia of Philosophy (Fall 2004 Edition), E. Zalta (ed.). (online).
• Lewis, David. (1966). "An Argument for the Identity Theory". Journal of Philosophy 63.
• Lewis, David. (1980). "Mad Pain and Martian Pain". In Block (1980a), Vol. 1, pp. 216–222.
• Lycan, W. (1987). Consciousness. Cambridge, MA: MIT Press.
• Mandik, Pete. (1998). Fine-grained Supervenience, Cognitive Neuroscience, and the Future of Functionalism.
• Marr, D. (1982). Vision: A Computational Approach. San Francisco: Freeman & Co.
• Polgar, T. D. (2008). "Functionalism". The Internet Encyclopedia of Philosophy.
• Putnam, Hilary. (1967). "Psychological Predicates". In Art, Mind, and Religion, W.H. Capitan and D.D. Merrill (eds.), pp. 37–48. (Later published as "The Nature of Mental States" in Putnam (1975a)).
• Putnam, Hilary. (1975a). Mind, Language, and Reality. Cambridge: CUP.
• Searle, John (1980). "Minds, Brains and Programs" (PDF). Behavioral and Brain Sciences. §56§ (3): 417–424. doi:10.1017/s0140525x00005756. S2CID 55303721.Philosophical Review LXVIII.

• Levin, Janet. "Functionalism". In Zalta, Edward N. (ed.). Stanford Encyclopedia of Philosophy. ISSN 1095-5054. OCLC 429049174.Eliasmith, Chris (2004). "Dictionary of the Philosophy of Mind - Functionalism". University of Waterloo (Canada). Archived from the original on 2012-06-23.Source: TORIma Academy Archive