Default Hypotheses in the Study of Perception

Jacob Berger[1] and
Myrto Mylopoulos[2]


A Reply to Phillips

Abstract: Some theorists have recently raised doubts about much of the experimental evidence purporting to demonstrate the existence of unconscious perception. In our 2019 article in this journal, we argued that some of these considerations are not decisive. Phillips (2021) replies thoughtfully to our paper, concluding that he is unconvinced by our arguments. Phillips maintains that the view that perception is invariably conscious remains, as he puts it, the ‘default’ hypothesis both within the folk understanding and experimental study of per­ception. There is much to agree with in Phillips’ piece, but there remain some substantive points of disagreement, which we outline here.

Keywords: perception; consciousness; problem of criterion; problem of attribution.

1. Introduction

Is there evidence for unconscious perception? Consider experiments involving masked priming, wherein target stimuli are visually presented for limited durations and masked by other stimuli, which results in participants reporting that they do not see the target stimuli, though the stimuli nonetheless prime downstream behaviour (e.g. Bachmann and Francis, 2013). A standard interpretation (for an over­view, see Persuh, 2018) is that while the target stimuli are not con­sciously perceived, they are unconsciously perceived. Call this hypothesis ‘Unconscious’.

There is, however, an alternative hypothesis — namely that, in many such studies, the target stimuli are consciously perceived, but in a degraded way, and participants thus fail to report perceiving them due to conservative response criteria. Call this competing hypothesis ‘Conscious’.

Phillips (e.g. Phillips, 2016; 2018; forthcoming; Phillips and Block, 2016; Peters et al., 2017) has raised doubts about much of the experi­mental evidence purporting to support Unconscious as against Con­scious. In our 2019 article, we argued that many of Phillips’, and some related, considerations are not decisive. Phillips (2021) replies thoughtfully to our paper, concluding that he ‘remain[s] unconvinced’ (p. 201).[3]

As we understand it, Phillips’ central claim in his reply is that Conscious is (or should be) what he calls the ‘default’ hypothesis both within the folk understanding and the experimental study of perception. It thus seemingly falls to proponents of Unconscious to provide compelling evidence for their position. In his work, however, Phillips carefully reviews much of the experimental literature, arguing that, for any result that proponents of Unconscious cite in favour of their view, they do not rule out two possibilities. First, most experi­ments simply do not control satisfactorily for criterion bias and thus do not rule out Conscious. This is often called ‘the problem of the criterion’. Second, many studies would seem vulnerable to what Phillips calls ‘the problem of attribution’: the possibility that target stimuli are not perceived at all, but rather only perceptually processed in a subpersonal way. Phillips concludes that Conscious thus remains the default hypothesis.

We begin in Section 2 by arguing that, despite Phillips’ misgivings, Unconscious may be the default hypothesis in folk and perceptual psychology. But even if Unconscious were not the default hypothesis, we propose that what matters is that Unconscious is an open hypothesis that is consistent with folk psychology and taken seriously by some cognitive scientists, which, as we argued in our 2019 paper, it plainly is. Moreover, we argue in Section 3 that as it stands there are no positive reasons to reject Unconscious and that more experimental work is thereby needed to settle the matter. In Section 4, we suggest some ways to move forward in this debate, urging that Phillips must provide positive reasons for thinking that Conscious is true, apart from its purportedly being the default hypothesis. We close in Section 5 by arguing that Phillips fails to adequately respond to the dilemma we posed in our 2019 article for the sceptic of Unconscious. We conclude that there remain good reasons to think that Unconscious not only may be, but also is, the correct hypothesis.

2. Unconscious as the ‘Default’ Hypothesis

In his reply, Phillips does not explicitly articulate what he takes a ‘default’ hypothesis to be, but his remarks suggest a few possible interpretations. First, he seems to indicate that a default hypothesis about some subject matter is the view that is largely accepted in both common-sense discourse and in the relevant science(s). In our 2019 paper, we made remarks suggesting that we regard Unconscious as the default hypothesis in this way, as we urged that rejecting Uncon­scious would entail revision of many individuals’ folk and scientific beliefs. Phillips, by contrast, contends that it is ‘certainly disputable’ that the folk or many experimentalists believe in unconscious per­ception (p. 192).

Starting with folk psychology, if Phillips were correct that it does not include unconscious perception, then perhaps Unconscious would be at a theoretical disadvantage. After all, to theorize about the nature of something, we must first have a common-sense grip on it, lest our theory fail to account for the relevant phenomenon (see Sellars, 1963). It is therefore important to correctly describe our ordinary understanding of perception at the outset.

In support of the view that common sense countenances uncon­scious sensory states, we cited, for example, Reuter and Sytsma’s (2020) work in experimental philosophy. In reply, Phillips raises some doubts about the methodology of those studies. But whether or not his criticisms are cogent, more basic evidence that there is folk-psychological support for Unconscious is that ordinary people and popular media often describe cases of unconscious perception such as subliminal advertising or hidden messaging (for a striking example, see Grow, 2015). Likewise, unconscious mentality has been widely discussed in the history of ideas. Prefiguring Freud, Nietzsche developed a complex account of the unconscious mind (see Katsafanas, 2005). And Coleridge marvelled that ‘Man exists… to himself & to God alone/ — Yea, in how much only to God — how much lies below his own Consciousness’ (1957, p. 1554). It is unclear to us why one would insist that perception is excluded in these con­ceptions of unconsciousness.

Phillips might reply that such uses of ‘perception’ nonetheless fail to distinguish between personal-level — and thus what he thinks is necessarily conscious — perception and subpersonal sensory pro­cessing. But we would need independent reason to think that this is the case. Moreover, even if this were correct, it would show only that the common-sense conception is not particularly nuanced and that even some theoretical conceptions of perception are not sensitive to that distinction. We thus think that Unconscious is part of folk psychology, or at least consistent with it, rather than a ‘heterodox’ view, as Phillips (p. 192) regards it.

Phillips argues instead that Conscious is the common-sense view, citing Campbell (2011) as insisting that our ordinary conception of seeing is of conscious seeing and noting that the OED defines per­ception as involving awareness (p. 192). But the OED’s definition is ambiguous between what Rosenthal (e.g. 2005, p. 4) has called ‘transitive’ consciousness or awareness — that is, one’s awareness of something — and so-called ‘state’ or ‘phenomenal’ consciousness, the sort of consciousness for which there is something that it is like to have it (see Block, 1995), which is the kind at issue here. It begs the question to assume that transitive consciousness cannot occur in the absence of state/phenomenal consciousness. And we can hardly credit Campbell’s assertion without further argument.

In any case, we regard it as somewhat of a scholastic point whether or not ordinary people, or historical thought, admit of unconscious perception. We think that they do, but we also think that folk views do not go very far in establishing what a theorist should think is actually the case. Even if the folk explicitly denied the possibility of uncon­scious perception, perhaps findings in perceptual psychology would nevertheless force us to amend our folk conception of perception to include it, just as findings in theoretical physics require us to revise our ordinary conceptions of matter, energy, and time.

What is more important is whether or not perceptual psychology takes Unconscious seriously as an hypothesis — and it’s clear that it does. As Kuhn (e.g. 1970, p. 12) urged, a reasonable way to determine if something is a relatively established phenomenon in a period of ‘normal science’ within a paradigm is whether or not it is taught in the textbooks. And unconscious perception is routinely mentioned in standard vision science and cognitive psychology texts (e.g. Palmer, 1999, pp. 639ff.; Sternberg and Sternberg, 2016, p. 95).

Further evidence that at least one corner of cognitive science takes Unconscious seriously is that, if it did not, what would theories of consciousness, such as the global neuronal workspace theory (e.g. Dehaene et al., 2006), purport to be theories of? Such views claim to explain the difference between conscious mental states and their unconscious counterparts — grounding it in differences such as availability for global broadcast. Theorists would not spill much ink exploring these views if they did not believe that there were uncon­scious personal-level perceptual states to contrast with conscious ones. One might think that such views are not accounts of the relevant kind of consciousness — that they are instead accounts of, for example, what Block (1995) calls ‘access consciousness’. But many such theories, including the global neuronal workspace theory, have been explicitly defended as accounts of state/phenomenal consciousness (see Naccache, 2018).

Phillips objects that we unfairly note in our 2019 paper only enthusiasm for unconscious perception among experimentalists, while not citing criticisms of it. We are happy to acknowledge such criticisms. But it goes without saying that in every scientific area there are disagreements regarding not only methodology, but also the nature and even the existence of target phenomena. While astrophysicists once widely accepted the presence of dark matter, for example, the search for it has turned up little; the alternative hypothesis that gravity itself operates differently at macroscales is one that some astro­physicists are now taking more seriously (for a popular overview, see Skibba, 2020). However this debate ultimately goes, that there are similar disputes about unconscious perception in cognitive science itself goes no way at all toward undermining the acceptance of Unconscious in the scientific community.

There are, however, passages in which Phillips hints at a slightly different notion of ‘default’ hypothesis, wherein an hypothesis is the default just in case it is ‘more parsimonious than its rivals’ (p. 191). Phillips maintains, moreover, that Conscious is simpler than Unconscious, endorsing Snodgrass’s view that Conscious ‘postulates only one rather than two [or more] perceptual processes’ (2002, p. 556, cited on p. 191). But Unconscious need not be, and is best not, construed as positing two perceptual processes. Instead, it should be viewed as positing a single perceptual process, which becomes conscious, via mechanisms described by the competing theories of consciousness, such as global broadcast.

Even if Unconscious were not the default hypothesis in either of the two ways scouted above, we do not think that this would make a significant difference to this debate. Alternative hypotheses are always available in any empirical domain — and we see no good reason to think that any special burden or status lies with less widely-held hypotheses than with more widely-held ones. Likewise, in general parsimony is a comparatively weak theoretical virtue (see Baker, 2016) — and arguably paradigm relative. Quantum mechanics is vastly more complex than Newtonian mechanics, but its complexity elegantly explains the available data.

Perhaps by ‘default hypothesis’ Phillips means instead something like the hypothesis that should be assumed within the target area of study unless alternative hypotheses can be proven — and thereby that the burden of proof lies with defenders of alternative hypotheses (see p. 186–7). But we doubt that this is the right way to think about hypotheses in general. Theory is always underdetermined by data. And, as good confirmation holists (e.g. Quine, 1951), we regard no hypotheses as confirmed or refuted by any restricted set of evidence, much less any particular finding. Indeed, a theoretical posit may be perfectly warranted, even in the absence of direct experimental evidence for it. Consider, for example, the long scientific search for experimental evidence to confirm the existence of the Higgs boson. Relatedly, some reasons to question evidence for an hypothesis do not require that we reject the hypothesis. It always remains open to adjust background assumptions to reject or reinterpret what may seem to be recalcitrant data, rather than give up on a theory. In light of Lavoisier’s findings about combustion, proponents of phlogiston attributed to it a negative weight, though the view was ultimately rejected in favour of the oxygen theory (e.g. Barker and Kitcher, 2014, pp. 82–4). While phlogiston theory was never — and arguably could never be — decisively refuted, the preponderance of evidence, coupled with various theoretical considerations, did the theory in.

Of course, no theorists can or should be expected to rule out all alternative hypotheses, given that many may be outlandish or not genuinely worth considering. Only genuinely open alternatives must be given weight and consideration. What makes an hypothesis ‘genuinely open’? While we do not venture a positive account, it is plainly not the mere fact that some theorists champion the hypothesis. Rather, it minimally has to do with whether scientists have good reasons to take the hypothesis seriously — such as that it explains much of the available data and makes fecund, precise, and testable predictions.

What matters is thus that Unconscious is an open hypothesis that is either endorsed by folk psychology, or at least consistent with it, and taken seriously by many cognitive scientists. We have argued that Unconscious meets these criteria. This was the main point of our 2019 paper, not that the view is the default hypothesis — whatever that may be.

What we want, then, are reasons to prefer either Conscious or Unconscious. We thus now review some of the ones we explored in our article, and consider Phillips’ replies.

3. Reasons to Reject Unconscious

In our 2019 paper, we discussed a series of experiments by Peters and Lau (2015) (‘P&L’), which seemed to us to supply the best evidence currently available in favour of Conscious as against Unconscious. P&L used a two-interval forced-choice (‘2IFC’) technique to find that participants who made judgments about the presence of pairs of masked stimuli were able to bet accurately on the judgment in which they were more confident. Although P&L interpreted their findings as supporting Conscious, we proposed that there is an explanation con­sistent with Unconscious, according to which participants do not consciously see the stimuli, but form conscious hunches about what was seen, which explain their betting behaviour. In his reply, Phillips agrees with us — though for independent reasons — that P&L’s study fails to offer novel support for Conscious.

Phillips argues that his own alternative hypothesis nevertheless remains preferable, once again urging that our Unconscious-plus-hunch explanation is less simple than Conscious. Our proposed explanation of P&L’s results purportedly posits not only two streams of perception, but also hunches, whereas Conscious posits only a single stream of conscious perception and response bias. However, we are doubtful that our proposed explanation really is more complex. Again, most theories of consciousness urge that the same token state can be conscious or not. And while such a consciousness mechanism may be psychologically complex, it is not more complex — theoretically speaking — than positing psychological mechanisms of response criteria. Similar remarks go for the positing of hunches, which we proposed can be understood as an ordinary form of meta­cognition evinced in this task.

For confirmation-holistic reasons, it is unsurprising that both Conscious and Unconscious can explain much of the same evidence. It is rarely the case that a false paradigm cannot offer some explanation of data. This is why we are not compelled by Phillips’ strategy of arguing that individual pieces of evidence put forth in favour of Unconscious are instead compatible with Conscious. By the same token, as we observed in our paper (2019, pp. 13–14), the possibility that participants in standard masking studies fail to report on masked stimuli due to conservative response bias is, when con­sidering most available studies, only speculative too. Without additional experimentation to tease these hypotheses apart, in many cases it remains open which hypothesis accurately describes what is going on.

Phillips is doubtless right that we therefore cannot conclude that much current evidence regarding phenomena such as masked priming or blindsight is decisive evidence for Unconscious. But no such studies are likewise evidence for Conscious, unless and until we control for criterion bias in the right way. (More on this shortly.) Assuming for the sake of argument that both Conscious and Uncon­scious equally fit the available data, perhaps we should not decide between them until more evidence is available.

4. Ways Forward?

We nonetheless do think there may be ways to move forward in this debate. After all, the hypotheses under consideration make different experimental predictions. The proponent of Conscious holds that, if we control for criterion bias, participants would verbally report what would otherwise seem to be unconscious perception. On Uncon­scious, by contrast, even if we adequately control for criterion bias there would be unreportable perceptual states.

To test between Conscious and Unconscious, we need a way to control not only for criterion bias, but also for the possibility that reporting effects are driven by conscious hunches and not conscious perception. The best attempt (so far) at controlling for criterion bias — namely, P&L’s experiments — do not offer unimpeachable results in favour of Conscious. But perhaps there is a kind of bias-free task that is also capable of controlling for the possibility of conscious hunches. If such an experiment demonstrated that, under such conditions, participants would verbally report on masked stimuli, then we would be more inclined to accept Conscious.

On the other hand, one might think that the most straightforward way to control for criterion bias is to ensure that perceptual discrimi­nation is at chance — that is, focus on experiments in which the signal-to-noise ratio is equal to 0, but priming effects are observed. But, as we observed in our article, such experiments purportedly face Phillips’ problem of attribution — namely, the worry that whatever priming effects participants demonstrate are driven not by personal-level unconscious perceptual states, but rather only by subpersonal states of perceptual processing.

Phillips’ argument here depends on adopting a rather stringent characterization of personal-level perception. On his view, the best evidence we have that a state is genuinely perceptual is that it is available to central coordinating agency or, as he more recently puts it, for the guidance of intentional action (e.g. Phillips, 2020, p. 299). But in the masked-priming experiments he canvasses in his 2018 paper (e.g. Jiang et al., 2006; Norman, Heywood and Kentridge, 2013), all we seem to have evidence for are ‘stimulus-driven, reflex-like’ (Phillips, 2018, p. 495) responses that fall short of the required form of agency.

We argued in our 2019 paper that, even granting Phillips’ criteria, we have evidence for unconscious visual states that play the role that Phillips requires of them. We focused especially on the case of DF, an individual with visual-form agnosia resulting from damage to her ventral stream. We urged on the basis of a discussion of Pisella and colleagues’ (2000) results that at slower timescales (over 300 ms) we have no reason to doubt that DF’s actions, which are guided solely by unconscious dorsal-stream vision, are sensitive to intention in the way Phillips’ criterion requires. We also argued that there is no basis to Phillips’ inference that the type of action that dorsal-stream vision supports is not attributable to the agent because it is automatic. As we stressed, much of our automatic behaviour — for example, that which is involved in skilled action — is so attributable, as evidenced by its sensitivity to and tight interfacing with an agent’s intentions (e.g. Fridland, 2017; Shepherd, 2019; Pacherie and Mylopoulos, 2020).

In response, Phillips argues on the basis of a different set of empirical results that DF’s actions may indeed be guided by conscious vision. He also argues against our interpretation of the empirical work we cite (Liu and Todorov, 2007) in favour of the sensitivity of auto­matic action to intention. There are, however, further studies to cite in response (e.g. Todorov and Jordan, 2002); and we suspect that there are ways to challenge Phillips’ interpretations of the relevant empirical results, which could be explored in future work.

But we again do not consider this kind of back-and-forth to be a particularly fruitful strategy for settling the debate. What we arguably need to do now is step back and examine theoretical considerations for and against these hypotheses.

Here is thus one theoretical worry for Phillips’ position: that his case rests on a distinction between the personal and subpersonal levels of perceptual processing that has been characterized in competing ways by many different theorists (for review, see Drayson, 2012) — and thus threatens to be arbitrary (e.g. Block’s contribution in Peters et al., 2017, p. 8).

Moreover, why should we think that there are any clean criteria by way of which we can draw this distinction? (Cf. Taylor, 2019, who thus concludes that there is no fact of the matter regarding when there is unconscious perception.) Indeed, we should expect our theorizing about the nature of personal-level perception to depend holistically on other theorizing regarding related phenomena such as action or perceptual thought (see Shepherd and Mylopoulos, forthcoming). Likewise, we should remain insistent that this debate be substantive. If we adopt overly demanding requirements on personal-level per­ception, perhaps not much hangs on whether we call complex uncon­scious perceptual processing ‘perception’.

Finally, we note that Phillips himself offers no positive theoretical reason favouring Conscious as against Unconscious, other than claiming that the former is the (simpler) default position. Such reasons should be given. For our part, we did offer one broadly theoretical argument against scepticism about unconscious perception in our 2019 article. We turn now to Phillips’ reply to that argument, which we find unsuccessful.

5. Revisiting the Dilemma for Scepticism
about Unconscious Perception

In our 2019 paper, we argued that sceptics about unconscious per­ception face a dilemma: either they must maintain that other kinds of mentality such as beliefs or desires also cannot occur unconsciously, or they must explain why perceptual states are unique in the mind in this regard. Both options, we maintained, are questionable.

Phillips attempts to resolve the dilemma by distinguishing mental states from mental occurrences, where the former do not unfold temporally and are thus not candidates for being conscious, while the latter do unfold temporally and can thus be conscious. Phillips seems to allow that mental episodes, such as the fading of an emotion, can occur without consciousness, but he maintains that these can be explained entirely by appeal to transitions between non-conscious mental states. He writes that, ‘In this light, it is not unreasonable to propose that all genuinely mental episodes other than those which can be wholly analysed in terms of transitions between non-conscious states are conscious’ (p. 201).

But we see no support for this position. Instead, we propose that some beliefs, desires, or emotional states arguably manifest in mental occurrences — thoughts, wishes, and so forth — that themselves can likewise occur without being conscious. And there is much support for this view. There is, for example, evidence that people can engage in unconscious inference. Consider Mandelbaum’s (2016) work on implicit bias, which demonstrates that we often act in biased ways towards individuals of certain social groups in ways that are best explained by our engaging in patterns of transitions between occurrent propositional attitudes of which we are unaware. As Mandelbaum argues, such inferences are not interactions among unconscious dispo­sitions, but among unconscious mental occurrences (see also Jenkin, 2020). Similar remarks go for other sorts of cognition, such as occurrent desiderative and emotional states (see Rosenthal, 2008). If this is right, then we can run the original dilemma again, but this time aimed at perception and other mental occurrences: why is it that perception alone always occurs consciously, while thoughts and wishes do not?

In reply, Phillips might appeal to the supposed temporal dimensions of perceptual occurrences that differ from those of their non-perceptual counterparts. But this cuts no ice — just as one can momentarily perceive red or have a perceptual experience of a passing scene that occurs over time, one can both momentarily have the thought that p or a fleeting desire — or think over time that p or long for something.

Without an independently motivated reason to think that thought and perception differ in a way relevant to consciousness, Phillips has not adequately addressed our dilemma. But we never intended the dilemma to be dispositive — simply one theoretical consideration to add to the collection of evidence that abductively supports the existence of unconscious perception.[4]


Many thanks to Richard Brown and David Pereplyotchik for their helpful comments on earlier drafts of this reply.


Bachmann, T. & Francis, G. (2013) Visual Masking: Studying Perception, Attention, and Consciousness, Oxford: Academic Press.

Baker, A. (2016) Simplicity, in Zalta, E.N. (ed.) The Stanford Encyclopedia of Philosophy, [Online], simplicity/.

Barker, G. & Kitcher, P. (2014) Philosophy of Science: A New Introduction, Oxford: Oxford University Press.

Berger, J. & Mylopoulos, M. (2019) On scepticism about unconscious perception, Journal of Consciousness Studies, 26 (11–12), pp. 8–32.

Block, N. (1995) On a confusion about a function of consciousness, Behavioral and Brain Sciences, 18 (2), pp. 227–247.

Campbell, J. (2011) Review essay on Origins of Objectivity by Tyler Burge, The Journal of Philosophy, 108 (5), pp. 269–285.

Coleridge, S.T. (1957) The Notebooks of Samuel Taylor Coleridge: Volume I, Coburn, K. (ed.), Princeton, NJ: Princeton University Press.

Dehaene, S., Changeux, J.-P., Naccache, L., Sackura, J. & Sergenta, C. (2006) Conscious, preconscious, and subliminal processing: A testable taxonomy, Trends in Cognitive Science. 10, pp. 204–211.

Drayson, Z. (2012) The uses and abuses of the personal/subpersonal distinction, Philosophical Perspectives, 26 (1), pp. 1–18.

Fridland, E. (2017) Skill and motor control: Intelligence all the way down, Philosophical Studies, 174 (6), pp. 1539–1560.

Grow, K. (2015) Judas Priest’s subliminal message trial: Rob Halford looks back, Rolling Stone, [Online], judas-priests-subliminal-message-trial-rob-halford-looks-back-57552/.

Jenkin, Z. (2020) The epistemic role of core cognition, Philosophical Review, 129 (2), pp. 251–298.

Jiang, Y.V., Costello, P., Fang, F., Huang, M. & He, S. (2006) A gender- and sexual orientation dependent spatial attentional effect of invisible images, Pro­ceedings of the National Academy of Sciences USA, 103 (45), pp. 17048–17052.

Katsafanas, P. (2005) Nietzsche’s theory of mind: Consciousness and con­ceptualization, European Journal of Philosophy, 13 (1), pp. 1–31.

Kuhn, T. (1970) The Structure of Scientific Revolutions, 2nd ed., with postscript, Chicago, IL: University of Chicago Press.

Liu, D. & Todorov, E. (2007) Evidence for the flexible sensorimotor strategies predicted by optimal feedback control, The Journal of Neuroscience, 27 (35), pp. 9354–9368.

Mandelbaum, E. (2016) Attitude, inference, association: On the propositional structure of implicit bias, Noûs, 50 (3), pp. 629–658.

Naccache, L. (2018) Why and how access consciousness can account for phenom­enal consciousness, Philosophical Transactions of the Royal Society B: Biologi­cal Sciences, 373 (1755), 20170357.

Norman, L.J., Heywood, C.A. & Kentridge, R.W. (2013) Object-based attention without awareness, Psychological Science, 24 (6), pp. 836–843.

Pacherie, E. & Mylopoulos, M. (2020) Beyond automaticity: The psychological complexity of skill, Topoi, [Online],

Palmer, S.K. (1999) Vision Science: Photons to Phenomenology, Cambridge, MA: MIT Press.

Persuh, M. (2018) The Fata Morgana of unconscious perception, Frontiers in Human Neuroscience, 12, art. 120.

Peters, M.A. & Lau, H. (2015) Human observers have optimal introspective access to perceptual processes even for visually masked stimuli, eLife, 4, e09651.

Peters, M.A., Kentridge, R., Phillips, I. & Block, N. (2017) Does unconscious perception really exist? Continuing the ASSC20 debate, Neuroscience of Con­sciousness 3 (1), nix015.

Phillips, I. (2016) Consciousness and criterion: On Block’s case for unconscious seeing, Philosophy and Phenomenological Research, 93 (2), pp. 419–451.

Phillips, I. (2018) Unconscious perception reconsidered, Analytic Philosophy, 59 (4), pp. 471–514.

Phillips, I. (2020) Object files and unconscious perception: A reply to Quilty-Dunn, Analysis, 80 (2), pp. 293–301.

Phillips, I. (2021) Scepticism about unconscious perception is the default hypothesis, Journal of Consciousness Studies, 28 (3–4).

Phillips, I. (forthcoming) Blindsight is qualitatively degraded conscious vision, Psychological Review, [Online], doi:10.1037/rev0000254.

Phillips, I. & Block, N. (2016) Debate on unconscious perception, in Nanay, B. (ed.) Current Controversies in Philosophy of Perception, pp. 165–192, New York: Routledge.

Pisella, L., Gréa, H., Tilikete, C., Vighetto, A., Desmurget, M., Rode, G., Boisson, D. & Rossetti, Y. (2000) An ‘automatic pilot’ for the hand in human posterior parietal cortex: Toward reinterpreting optic ataxia, Nature Neuroscience, 3 (7), pp. 729–736.

Quine, W.V. (1951) Two dogmas of empiricism, Philosophical Review, 60, pp. 20–43.

Reuter, K. & Sytsma, J. (2020) Unfelt pain, Synthese, 197 (4), pp. 1777–1801.

Rosenthal, D. (2005) Consciousness and Mind, Oxford: Clarendon Press.

Rosenthal, D. (2008) Consciousness and its function, Neuropsychologia, 46 (3), pp. 829–840.

Sellars, W. (1963) Science, Perception, and Reality, London: Routledge and Kegan Paul.

Shepherd, J. (2019) Skilled action and the double life of intention, Philosophy and Phenomenological Research, 98 (2), pp. 286–305.

Shepherd, J. & Mylopoulos, M. (forthcoming) Unconscious perception and central coordinating agency, Philosophical Studies.

Skibba, R. (2020) Does dark matter exist?, Aeon, [Online], why-its-time-to-take-alternatives-to-dark-matter-seriously.

Snodgrass, M. (2002) Disambiguating conscious and unconscious influences: Do exclusion paradigms demonstrate unconscious perception?, American Journal of Psychology, 115 (4), pp. 545–579.

Sternberg, R.J. & Sternberg, K. (2016) Cognitive Psychology, 6th ed., Boston, MA: Wadsworth Cengage Learning.

Taylor, H. (2019) Fuzziness in the mind: Can perception be unconscious?, Philosophy and Phenomenological Research, [Online], phpr.12592.

Todorov, E. & Jordan, M.I. (2002) Optimal feedback control as a theory of motor coordination, Nature Neuroscience, 5 (11), pp. 1226–1235.

Paper received September 2020; revised October 2020.

[1]      Department of Philosophy, Lycoming College, Williamsport, PA 17701, USA.

[2]      Department of Philosophy and Department of Cognitive Science, Carleton University, Ottawa, ON K1S 5B6, Canada.

[3]      All references to Phillips’ work refer to his 2021 paper in this journal, unless otherwise noted.

[4]      Managing editor’s note: JCS will not be considering further papers concerning this particular debate. However, this article will be uploaded to the JCS blog following publication (, where the authors concerned (and any other interested readers) can continue to discuss the topic.

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Stay Informed, Stay Inspired

Discover the frontier of academic insights with our newsletter. Imprint Academic brings a world of scholarly discourse right to your inbox. Stay updated on the latest publications, special offers, and the vibrant conversations shaping today’s academic landscape. Your quest for knowledge deserves a companion as dedicated as our newsletter. Sign up now and become a part of a community driven by curiosity and intellectual exploration.
Something went wrong. Please check your entries and try again.