CC BY
32
I HENOMENAL CHARACTER, REPRESENTATIONAL CONTENT, AND THE INTERNAL CORRELATION OF EXPERIENCE: ARGUMENTS AGAINST TRACKING REPRESENTATIONALISM
Tracking representationalism is the theory that phenomenal consciousness is a matter of tracking physical properties in an appropriate way. This theory holds that phenomenal character can be explained in terms of representational content, and it also entails that there is unlikely to be a strong correlation between phenomenal character and neural states. However, the empirical evidence shows that both claims cannot be true. So, tracking representationalism is wrong. Its fault is due to ignoring the internal correlation of experience, the existence of which shows that phenomenal character is shaped by neural states to a large extent, so it cannot be wholly explained by representational content.
Keywords: tracking representationalism; phenomenal character; representational content; the internal correlation of experience.
I. Introduction
Tracking representationalism is a contemporary theory of the phenomenal character of experience. To be specific, this theory holds that experiences are states representing the world in an appropriate way, and their phenomenal character can be explained by the representational contents. In addition, how experiences obtain their representational contents can be explained in terms of a tracking relation to physical properties [Bourget and Mendelovici, 2014: 210]. If this theory is correct, then we are able to put forward a physicalist account of phenomenal character and provide a simple answer to the "hard problem" of explaining why there is phenomenal character for a subject in experience [Chalmers, 1995: 201-202]. That is why it attracted a great deal of proponents, including Fred Dretske, William Lycan, and Michael Tye, in the recent years.
In this paper, I argue against tracking representationalism by using empirical evidence as well as a priori methods. My paper assumes the following structure. In section II, I sketch tracking representationalism and the motivation for it - the naturalization of phenomenal character. In section III, I put forward my first argument, according to which there is little correlation between phenomenal character and representational content, so the core thesis of tracking representationalism thatpheno-menal character can be explained in terms of representational content is untenable. In section IV, I put forward my second argument, according to which the internal correlation (viz., the isomorphic relationship between the structure of phenomenal
Bin Zhao - Department of Philosophy, Shandong University E-mail:
zhaobin529969263@ gmail.com
218 Interdisciplinary Studies
character and the structure of neural states) does exist, but that conflicts with the implication of tracking representationalism that there is unlikely to be such a correlation between phenomenal character and neural states. Finally, in section V, I elaborate that tracking representationalists' failure is due to their ignorance of the internal correlation. The existence of the internal correlation shows that phenomenal character is shaped by neural states to a large extent, so it cannot be wholly explained in terms of representational content.
II. Tracking Representationalism and the Naturalization of Phenomenal Character
Tracking representationalism is a theory ofphenomenal consciousness, the "what it is like" of experience. It aims to understand consciousness in terms of intentionality. Roughly speaking, intentionality is the aboutness or directedness of mental states. For example, we can believe that water is H2O rather than XYZ, or desire that it will rain tomorrow. These mental states both exhibit a kind of aboutness or directedness to the world. What a mental state is about or directed to is its representational content.
At a minimum, representationalism holds that experiences with the same representational content have the same phenomenal character and experiences with different phenomenal character have different representational contents. That is to say, phenomenal character supervenes on representational content. This fits well with our common sense. Suppose you are looking at a red rose, your experience of this rose has a particular phenomenal character. For the sake of simplicity, I will call this phenomenal character "reddish phenomenal character". And your experience also has a particular representational content. If the color of the rose changes and the phenomenal character of your experience changes with it, then your experience will also have a different representational content [Bourget and Mendelovici, 2014: 219]. Nevertheless, why is the difference in phenomenal character accompa- ^ nied by a difference in representational content? A rough explanation of this tf) observation is that phenomenal character is a specific representational content or representational content meets certain further conditions [Tye, 2002: IB 137]. For example, in the above case, your experience has a reddish pheno- ■£ menal character in virtue of the fact that it represents the sensible quality red, in other words, in virtue of the fact that the sensible quality red enters into the JjJ representational content of your experience.1 iS
1 Besides sensible qualities, the shape, size, location, and spatial relationships of external objects also enter into the representational content. But that has little relevance to our discussion.
Since experiencing a reddish phenomenal character is just representing the sensible quality red, what are sensible qualities? As a reductive version of representationalism, tracking representationalism advocates a physicalist account of sensible qualities, according to which sensible qualities are response-independent physical properties. For example, colors are surface reflectance, smell qualities and taste qualities are chemical properties of odorants and foods, auditory qualities are constituted by physical properties involving frequency, amplitude, and wave shape, and pain is tissue damage [Pautz, 2014a: 239]. In short, tracking representationalism holds that sensible qualities which constitute representational content are physical properties. And because phenomenal character can be explained by the representational content. Thus experiencing a specific phenomenal character is just representing a specific physical property or a complex of physical properties.
In addition, tracking representationalism advocates the tracking theory of intentionality, according to which, "intentionality is a matter of. . . correlating with features of the environment" [Bourget and Mendelovici, 2014: 215]. In detail, as Cutter and Tye illustrate:
[A] state represents a property or state of affairs in virtue of causally co-varying with that property or state of affairs under typical or optimal conditions.... Tokens of a state S in an individual x represent thatp in virtue of the fact that: under optimal conditions, x tokens S iffp, and because p [Cutter and Tye, 2011: 90-91, emphasis in original].
That is to say, representation relation (the relation between representational vehicle and the property or state of affairs being represented) can be understood in functionalist terms. Therefore, the fact that experience represents physical properties is demystified.
In short, tracking representationalism is the combination of the following two doctrines:
Representationalism: The phenomenal character of an experience (at a minimum) supervenes on its representational content, and the former can be explained by the latter.
Tracking theory of intentionality: A state represents a property or state of affairs in virtue of causally co-varying with that property or state of affairs under optimal conditions [Cutter and Tye, 2011: 90].
If this theory is correct, then we can provide an easy answer to the hard problem. The hard problem is the problem of explaining why an experience should have the specific phenomenal character it does, rather than having !S some other phenomenal character or having no phenomenal character at all. For example, when you see something red, you usually experience a И reddish phenomenal character. In virtue of what does your experience have 4 the reddish phenomenal character rather than having a throbbing painful-ф ness or having no phenomenal character at all? Chalmers argues that functional explanation does not apply to the hard problem, because "it is not a problem about the performance of functions. The problem persists
И
even when the performance of all the relevant functions is explained" [Chalmers, 1995: 201]. Moreover, we ordinarily suppose that perceptual experience is in the brain. But in the brain, we can only find grey matter and white matter, and there is no place for us to find phenomenal character. Where is phenomenal character located? Tracking representationalism provides a promising answer to these problems. According to this theory, phenomenal character is not an intrinsic property of experience, so we cannot find it in the brain. In fact, it is a representational property of experience. It can be explained in terms of the representational content which is constituted by physical properties represented by the experience. Moreover, on tracking representationalism, representation relation can be explained in functionalist terms. Therefore, the fact that there is "something it is like" for a subject in experience is no longer a myth.
In brief, tracking representationalism is a two-step strategy of the naturalization of mind. The first step - representationalism - accounts for phenomenal character in terms of representational content. The second step -tracking theory of intentionality - accounts for representational content in terms of a tracking relation to the physical properties [Bourget and Mende-lovici, 2014: 210]. Therefore, phenomenal character is naturalized and the hard problem is solved.
III. Phenomenal Character and
Representational Content
According to tracking representationalism, the phenomenal character of an experience can be explained by its representational content. This radical idea aroused considerable controversy. Philosophers constructed many thought experiments to refute it such as "the brain in vat", "swampman", "inverted spectrum", and "inverted earth".2 What they have in common is denying that phenomenal character supervenes on representational content. Thus, denying that the former can be explained by the latter. For example, in the thought experiment "inverted spectrum", visual experiences of two people can represent the same surface reflectance, while one has a reddish phenomenal character, the other has a greenish phenomenal character [Shoemaker, 1994: 34]. So phenomenal character does not supervene on representational content and, therefore, the former cannot be explained by the latter. In tracking representationalists' view, although scenarios in these thought experiments are conceivable, they are metaphysically impossible. It is just like the case where we can conceive gold is not element 79, but that is impossible [Tye, 1986: 5-6]. In brief, the intuition these
И
w
1Л
2 See: Horgan and Tienson (2002), Block (1998), Shoemaker (1994), and Block (1990) for related discussions.
thought experiments rely on is controversial, so conclusions drawn from them are unreliable.
Unlike these thought experiments, I will argue against tracking represen-tationalism by using empirical evidence in this section. Contrary to tracking representationalism, empirical evidence shows that there is little correlation between phenomenal character and representational content, so phenomenal character cannot be explained in terms of representational content. Thus empirical evidence poses a challenge to tracking representationalism. I will use olfactory experience as an example to illustrate this challenge in detail.3
When we smell a smell, we usually experience some phenomenal character. On tracking representationalism, these phenomenal character is determined by and can be explained by the representational content of experience. But research shows that there is little correlation between phenomenal character and representational content, so we are not able to explain phenomenal character in terms of representational content. For example, although hydrogen sulfide (molecular formula: H2S) and borane (molecular formula: BH3) both smell like rotten eggs, they have totally different molecular structures, in this case, two experiences have very similar phenomenal character, but their representational contents are totally different;4 although L-carvone and D-carvone have very similar molecular structures (they are stereisomers -molecules with exactly the same molecular formula and slightly different structures, just like left hand and right hand), they smell very different - the former smells like spearmint, while the latter smells like caraway [Da Costa and Eri, 2009:27], in this case, two experiences have totally different phenomenal character, but their representational contents are very similar. Since phenomenal character cannot be explained by representational content, tracking representationalism is refuted directly.
To this refutation, tracking representationalists have several possible replies. The first reply says what is represented by the olfactory experience is some chemical property other than the structure of odor molecule. The second reply says that every odor is a disjunction of some odor molecules rather than a specific structure of odor molecule, and it is this disjunction that is represented by the olfactory experience [Lycan, 2014: 7]. In what follows, I will argue these two replies cannot save tracking representatio-в) nalism in the face of my challenge.
IH 3 For the sake of simplicity, I only focus on olfactory experience in this section, though
C the thesis that phenomenal character cannot be explained in terms of representational content
!S applies to experiences of all sensory modalities such as vision, audition, and gustation. See
Pautz (2014a) and Goldstein (2010) for further discussions. JjJ 4 Though many tracking representationalists maintain that olfactory experience
■S represents qualities of odorants, for example, Tye (2002), as far as I know, none of them have
^ developed a detailed view (Batty (2009) may be an exception. Nonetheless, I do not know
^ whether she is a tracking representationalist, though she has developed a representational
account of olfactory experience). For the purpose of discussion, I assume that olfactory ^¡^ experience represents the structure of odor molecule in this passage. This may be the most promising view on the side of tracking representationalists.
Now let us turn to the first reply, according to which what is represented by the olfactory experience is some chemical property other than the structure of odor molecule. So which kind of chemical property is the "suitable candidate"? Tracking representationalists may turn to the vibrational theory of olfaction to defend their theory. According to the vibrational theory, what is perceived by the olfactory experience is the vibration frequency of chemical bond of odor molecule. This theory can explain the fact that hydrogen sulfide and borane have the same odor though their molecular structures are totally different, and this is because the S-H bond of hydrogen sulfide and the B-H bond of borane have very close vibration frequencies [Turin, 2009: 261]. So olfactory experiences of these two odor molecules have very similar representational contents, no wonder they have very similar phenomenal character.
Nevertheless, the explanatory power of the vibrational theory of olfaction is very limited, as there are many facts which cannot be explained by it. For instance, it cannot explain the case of carvone mentioned above. L-car-vone and D-carvone not only have very similar molecular structures, but also their corresponding chemical bonds have exactly the same vibration frequencies, however they smell different. In this case, two olfactory experiences have totally different phenomenal character, but their representational contents are the same.
Moreover, in some cases, the change in the concentration of an odor molecule can alter the phenomenal character of olfactory experience. For example, "the odor of thioterpineol is described as 'tropical fruit' at a low concentration, as 'grapefruit' at a higher concentration, and as 'stench' at a still higher concentration" [Pautz, 2014a: 245]. However, concentration cannot alter the vibration frequencies of chemical bonds. In this case, we have olfactory experiences which have different phenomenal character and the same representational content. Therefore, phenomenal character cannot be explained by representational content.
According to the second reply, every odor is a disjunction of some odor molecules rather than a specific structure of odor molecule. For example, the odor "smells like rotten eggs" can be characterized as "hydrogen sulfide v borane v sulfur monoxide v ... (this disjunction may be infinite)", just as some tracking representationalists hold the idea that color is a disjunctive microstructural property of objects [Lycan, 1996: 73]. Inaddition, it is this disjunction that is represented by the olfactory experi- k ence. So in the case of hydrogen sulfide and borane, two olfactory experiences represent the same disjunction, thus they have the same representa- !S tional content. That's why they also have the same phenomenal character.
However, this reply cannot eliminate all the counterexamples. For ins- in tance, "androstadienone ... is perceived by different individuals as offen- ^ sive ('sweaty, urinous'), pleasant ('sweet, floral') or odourless" [Keller et Q al., 2007: 468]. If the case is "androstadienone is perceived by different individuals as offensive or odorless", then tracking representationalists could
(A
explain it. Their explanation will be that the first group of people represented the disjunction "androstadienone v ...", so they smelled something offensive; while the latter group of people failed to represent this disjunction due to some dysfunction, so they did not smell anything.
Nonetheless, this explanation does not work for the case here, because there is another group of people who smelled something pleasant. What is represented by them? Firstly, we have no reason to assume that these people were in states of misrepresentation, viz., representing the odor molecule as having a property which it does not possess, because the test is conducted under normal conditions and the subjects do not have physiological or cognitive defects. Secondly, we have no reason to suppose that they represented other odor molecules or another disjunction of odor molecules because only pure androstadienone is used in this test. Therefore, even resorting to the disjunction of odor molecules cannot save tracking represen-tationalsim from my argument.
The above analysis shows that there is little correlation between phenomenal character and representational content. Thus, the core thesis of tracking representationalism that phenomenal character can be explained in terms of representational content is untenable, even resorting to ad hoc hypotheses like "every odor is a disjunction of some odor molecules" cannot defend it.
IV. The Internal Correlation of Experience
In section II, I have mentioned that tracking representationalism advocates the tracking theory of intentionality, according to which a state represents another state in virtue of the fact that there is a causal covariant relation between them rather than they have some properties in common. For instance, we can hardly say that the word "cup" and a cup have anything in common, but the former can still represent the latter. On tracking representationalism, this idea also applies to experiences such as pain. As the representational vehicle, the neural correlate of a pain experience represents a tissue damage of a specific intensity in virtue of the fact that the former is caused by the latter under optimal conditions rather than they have some properties in common.5 Even if the neural state did not have the properties which it actually has, for example, if it had a different firing rate, it can still represent the tissue damage of a specific intensity as long as there is an ap-y propriate causal covariant relation between it and the tissue damage.
Therefore, the firing rates of neural states are unlikely in direct proportion to the intensities of tissue damage. In addition, tracking representatio-
V)
1Л
5 In this paper, by "pain" I mean the sensory component of pain rather than the affective component of pain. For related discussions, see Goldstein (2010).
nalism holds that pain intensity is determined by the intensity of tissue damage represented by the pain experience, so pain intensities are unlikely in direct proportion to the firing rates of neural states [Pautz, 2014b: 308]. That is to say, there is unlikely to be an isomorphic relationship between the structure of phenomenal character (viz., the similarities or numerical relationships between phenomenal character) of pain experience and the structure of neural states (viz., the similarities or numerical relationships between neural states). For the sake of simplicity, call this isomorphic relationship "the internal correlation" [Pautz, 2014a: 242]. In sum, tracking representationalism implies that the internal correlation is not possible to exist.
However, the above implication conflicts with neuroscientists' discovery that pain intensity closely matches firing rates of neurons in the primary somatosensory cortex [Timmermann et al., 2001: 1501]. So the internal correlation does exist. In addition, neuroscience research shows that this thesis applies to experiences of all sensory modalities such as vision, olfaction, gustation, and audition [Pautz, 2014a; Goldstein, 2010]. Thus, contrary to the implication of tracking representationalism, there is a strong correlation between the phenomenal character of experience and neural states.
A possible response of tracking representationalists is that the internal correlation is just a coincidence because it does not exist in some counter-factual cases. Its reason is as follows: What an experience normally represents is what nature designed it to represent, what it has as its biological purpose to represent [Tye, 1998: 461]. For instance, we have reddish phenomenal character in some cases because some of our brain states have the biological function of representing red objects. And what biological function we have depends on our evolutionary history [Dretske, 1995: 7]. If our evolutionary history was slightly different, these brain states could be designed to represent blue objects rather than red objects. Thus, even if two individuals have the same internal states, their internal states can still represent different physical properties if the species they belong to have different evolutionary histories [Tye, 2009: 195].6 Likewise, neural states ofthe same firing rate can represent tissue damages of different pain intensities if their subjects have different evolutionary histories. In the real case, the nature accidentally designed a neural state whose firing rate is n units to represent tissue damage whose intensity is kn units (k is a constant). This leads to the fact that pain intensities are in direct proportion to the firing rates of neural states. While in some counterfactual cases, the nature designed a neural state whose firing rate is n units to represent tissue damage whose intensity is not kn units. Then pain intensities are not in direct proportion to the firing rates of neural states. So pain intensities are in direct proportion
6 Here is a simple analogy: the same voltmeter can perform different functions under different designs. It can be designed as a speedometer, then to represent the speed of a car; it also can be designed as a fuel gauge, then to represent the volume of oil. In these two cases, the internal states of the voltmeters are the same, but what they represent are different. For detailed discussions; see: Tye (2009).
И
W
1Л
to the firing rates of neural states in some cases is not attributable to the fact that there is a strong correlation between them. Their ostensible strong correlation is accidentally caused by the design on the earth.
The above response tries to explain away the correlation between phenomenal character and neural states by resorting to evolutionary history. I will use a thought experiment to argue that this response implies that phenomenal character has no causal effects on the physical. But there is some tension between this implication and some tracking representationalists' claim that "phenomenal character... [can] make a direct impact on the beliefs/desires system" [Tye, 1995:138].7 Thus the above response fails as a suitable response of tracking representationalists. My thought experiment is as follows:8
Oscar is an ordinary man, and Twin Oscar is his microphysical duplicate who was created in a chemical reaction caused by a lightning. At time t, they were respectively transported to planet A and planet B which are suitable for human habitation. Suppose there are the same physical events on planet A and planet B at time t, and these two planets obey the same physical laws. If causal closure applies to the physical world, then in the following time, planet A and planet B will be identical in all physical aspects.9 As an ordinary man, Oscar can experience phenomenal character. While Twin Oscar does not have an evolutionary history and nature did not make any design on him. So his internal states cannot play the role of representation. As a consequence, he cannot experience phenomenal character. Thus there are the same physical events and different phenomenal character on planet A and planet B after time t. The best explanation to this conclusion is that phenomenal character has no causal effects on the physical.
This implication totally violates our common sense. And, as I mentioned above, it is also in tension with some tracking representationalists' claim that phenomenal character has a direct impact on beliefs (or desires) [Tye, 1995:138].10 (For example, when we experience a reddish phenomenal character under normal conditions, we usually form a belief that we have seen something red. So the reddish phenomenal character (or the experience which has a reddish phenomenal character) has causal effects on our beliefs.) Thus the above response, which aims at explaining away the
^ 7 This claim aims to demarcate between representation with phenomenal character (for
И example, experience) and representation without phenomenal character (for example, desire
and belief). See: Tye (1995) for related discussions.
м 8 This thought experiment derives from the "swampman" thought experiment in
В Davidson (1987).
!S 9 The influence external environment has on these two planets is ignored, as it has little
Л relevance to the discussion here.
■M
JjJ 10 There is a possible solution to dissolve the tension. This solution maintains that though
■S phenomenal character (or the experience which has a phenomenal character) has causal
4 effects on beliefs (or desires), while beliefs and desires do not have causal effects on the
ф physical, so phenomenal character (or the experience which has a phenomenal character) does
not have (direct or indirect) causal effects on the physical. However, this solution also violates our common sense. As far as I know, tracking representationalists have not advocated such a strange view on belief and desire.
internal correlation, fails as a suitable response of tracking representationalists. This means that tracking representationalists are unable to explain away the internal correlation which is incompatible with their theory. So tracking representationalism is in trouble.
V. Where Tracking Representationalism Went Wrong
In the last two sections, I argued that (a) phenomenal character cannot be explained in terms of representational content, and (b) there is a strong correlation between phenomenal character and neural states, which conflicts with the implication of tracking representationalism, so tracking re-presentationalism fails. In my view, its failure is due to ignoring the internal correlation mentioned above, the existence of which shows that phenomenal character is shaped by neural states to a large extent, so it cannot be wholly explained in terms of representational content.
Take color experience for instance, opponent process theory of color vision holds that the phenomenal character of color experience can be explained by activation levels of opponent processing channels. Opponent processing channels consist of red-green opponent channel and yellow-blue opponent channel. When red-green opponent channel assumes a positive state of activation, we have a reddish experience; when it assumes a negative state of activation, we have a greenish experience. When yellow-blue opponent channel assumes a positive state of activation we have a yellowish experience; when it assumes a negative state of activation we have a bluish experience. Thus, according to opponent process theory, the phenomenal character of color experience is determined by activation levels of red-green opponent channel and yellow-blue opponent channel. Though just resorting to representation of surface reflectance cannot explain why there does not exist a binary color combined of red and green or a binary color combined of yellow and blue, opponent process theory can explain why these binary colors do not exist. This is because neither of the two channels can be in a state of both activation and inhibition at the same time [Pautz, 2006].
Thus it can be seen that though our color experience is a representation
И
of physical properties (surface reflectance), its phenomenal character cannot be wholly explained by representational content. On the contrary, many features of phenomenal character (for instance, the structure of phenome- !S nal character of color experiences) can only be explained by the internal states (for instance, the activation levels of opponent processing channels), in which shows that internal states play an important role in determining the ^ phenomenal character of experience [Pautz, 2006: 228]. g
In sum, experience is like footprints in the snow. The formation of footprints is not only related to the texture of snow, but also related to the shape =
of foot; which phenomenal character an experience has is not only related to its represnetational content, but also related to the internal states of the individual. In other words, the phenomenal character of experience is the product of the interaction between internal factors and external factors. Tracking representationalism duly appreciates the representational character of experience, but not the internal correlation of experience. This leads to the mistaken ideas that the function of neural states is just representation and phenomenal character can be explained by representational content. However, in fact, neural states shape our phenomenal character to a large extent, so the relationship between them cannot be the relationship between representational vehicle and representational content. Thus, we are not able to explain phenomenal character in terms of representational content.
И
VI. Conclusion
Tracking representationalism aims to put forward a naturalistic account of phenomenal character. If this theory is correct, then phenomenal character is determined by and can be explained by representational content which in turn can be explained in terms of a tracking relation to physical properties. Given my arguments, tracking representationalism is confronted with insurmountable problems. It cannot defend its core thesis that the phenomenal character of an experience can be explained by its representational content, and it also conflicts with neuroscientists's discovery that there is a strong correlation between phenomenal character and neural states (viz., the internal correlation). So tracking representationalism fails as a suitable naturalistic theory of mind. Its fault is due to ignoring the internal correlation of experience, the existence of which shows that phenomenal character is shaped by neural states to a large extent, so it cannot be wholly explained by representational content. Thus we should give up tracking representationalism and seek a theory which is compatible with the internal correlation.
References
™ Batty, Clare. 2009. "What's That Smell?" Southern Journal of Philosophy, 47
a& (4): 321-348.
W Block, Ned. 1990. "Inverted Earth." Philosophical Perspectives, 4: 53-79.
« —. 1998. "Is Experience Just Representing?" Philosophy and Phenomenolo-
gical Research, 58 (3): 663-670. A Bourget, David and Angela Mendelovici. 2014. "Tracking Representational-
ism: William Lycan, Fred Dretske, and Michael Tye." In Andrew Bailey (ed.), Phi-3 losophy of Mind: The Key Thinkers. London: Bloomsbury Publishing.
Chalmers, David J. 1995. "Facing Up to the Problem of Consciousness." Journal of Consciousness Studies, 2 (3): 200-219.
Cutter, Brian and Michael Tye. 2011. "Tracking Representationalism and the Painfulness of Pain." Philosophical Issues, 21 (1): 90-109.
Da Costa, Neil C. and Sanja Eri. 2009. "Identification of Aroma Chemicals." In David J. Rowe (ed.), Chemistry and Technology of Flavors and Fragrances. Oxford: Blackwell Publishing.
Davidson, Donald. 1987. "Knowing One's Own Mind." Proceedings of the American Philosophical Association, 60 (3): 441-458.
Dretske, Fred. 1995. Naturalizing the Mind. Cambridge, MA: MIT Press.
Goldstein, E. Bruce. 2010. Sensation and Perception. Belmont, CA: Wadsworth.
Horgan, Terence and John Tienson. 2002. "The Intentionality of Phenomenology and the Phenomenology of Intentionality." In David J. Chalmers (ed.), Philosophy of Mind: Classical and Contemporary Readings. Oxford: Oxford University Press.
Keller, Andreas, Hanyi Zhuang, Qiuyi Chi, Leslie B. Vosshall, and Hiroaki Matsunami. 2007. "Genetic Variation in a Human Odorant Receptor Alters Odour Perception." Nature, 449 (7161): 468-472.
Lycan, William. 1996. Consciousness and Experience. Cambridge, MA: MIT Press.
—. 2014. "The Intentionality of Smell." Frontiers in Psychology, 5: 1-8.
Pautz, Adam. 2006. "Sensory Awareness is not a Wide Physical Relation: An Empirical Argument Against Externalist Intentionalism." Nous, 40 (2): 205-240.
—. 2014a. "The Real Trouble with Phenomenal Externalism: New Empirical Evidence for a Brain-Based Theory of Consciousness." In Richard Brown (ed.), Consciousness Inside and Out: Phenomenology, Neuroscience, and the Nature of Experience. Dordrecht: Springer Netherlands.
—. 2014b. "Ignoring the Real Problems for Phenomenal Externalism: A Reply to Hilbert and Klein." In Richard Brown (ed.), Consciousness Inside and Out: Phenomenology, Neuroscience, and the Nature ofExperience. Dordrecht: Springer Netherlands.
Shoemaker, Sydney. 1994. "Phenomenal Character." Nous, 28: 21-38.
Timmermann, Lars, Markus Ploner, Katrin Haucke, Frank Schmitz, Rudiger Baltissen, and Alfons Schnitzle. 2001. "Differential Coding ofPain Intensity in the Human Primary and Secondary Somatosensory Cortex." Journal of Neurophysiology, 86: 1499-1503.
Turin, Luca. 2009. "Rational Odorant Design." In David J. Rowe (ed.), Chemistry and Technology of Flavors and Fragrances. Oxford: Blackwell Publishing.
Tye, Michael. 1986. "The Subjective Qualities of Experience." Mind, 95 Ц) (377), 1-17.
—. 1995. Ten Problems of Consciousness. Cambridge, MA: MIT Press.
—. 1998. "Inverted Earth, Swampman, and Representationalism." Philosophical Perspectives, 32:459-78.
—. 2002. "Representationalism and the Transparency of Experience." Nous, 36 (1): 137-151.
—. 2009. Consciousness Revisited: Materialism without Phenomenal Concepts. Cambridge, MA: MIT Press.
