This paper seeks to provide a rigorous definition of the concept of Cognitive Penetration (Pylyshyn 1999) or, as it is nowadays, more often called Cognitive Permeation (CP) (Burnston 2017). Alleged cases of CP include the perception of brightness in faces based on beliefs about their race and skin colour (Macpherson 2012), the perception of colour based on beliefs about the typical colour of objects (Witzel 2016), and the perception of facial features, based on beliefs about their emotions (Marchi & Newen 2015). The debate whether perception is permeated by cognition is ongoing (Toribio 2015, Firestone and Scholl 2015, Newen & Vetter 2017). One of the reasons is that we do not have a sufficiently precise definition of CP

In the literature, there are two approaches to understanding CP. On the one hand, the neurobiological approach conceptualises CP as a (sufficiently direct) causal influence on a specific segment of visual processing known as Early Vision. According to this approach, whether perception is permeated by cognition depends on whether there are direct neuronal links from cognitive segments to early vision segments. Approaches from philosophy and perceptual psychology, on the other hand, focus primarily on the kinds of mental representations involved and on whether the influence from cognitive states on perceptual states, can be ascribed to the content of these states or other contributors such as attention, etc. This paper argues that the latter approach to defining CP is superior. Only this approach to defining CP is relevant for theories of perceptual justification (Siegel 2012).

Based on the selected approach then, the study commences with defending four criteria of adequacy for a definition of CP.

1. The causal connection between the cognitive and perceptual mental representations is internal and mental.

2. There must be a semantic connection between the cognitive mental representation and the perceptual mental representation.

3. The causal connection is due to a computation over the semantic content of the cognitive and perceptual mental representations.

4. CP is not Theory-Ladenness of Perception.

These criteria are motivated by discussing examples from the literature (Macpherson 2017, Stokes 2015) of causal influence from cognition to perception. Each of these examples also shows that previous proposal for defining CP are inadequate because they do not satisfy all of the above criteria of adequacy for a definition of CP.

This discussion of the criteria of adequacy ends with a dilemma that explains why previous definitions of CP were inadequate. If the content of perceptual mental representation were identical to the conceptual content of belief-like cognitive mental representations, then CP and the theory-ladenness of perception could not be distinguished sufficiently. Suppose the contents of perceptual mental representation were sufficiently different from the conceptual contents of beliefs, i.e., by being non-conceptual, as many philosophers of perception suggest. In that case, previous approaches to CP and the philosophy of perception generally cannot explain how a semantic relationship between belief-like cognitive mental representations and non-belief-like perceptual mental representations can exist.

To resolve this dilemma and to offer an adequate definition of CP, the paper provides an account of the non-conceptual content of perceptual mental representations and its semantic relations to the conceptual content of beliefs. The account is a generalisation of my previous account of the semantic relationships between perception and beliefs (omitted) which is also applied to understanding the interface between the non-conceptual content of sensorimotor representations and the conceptual content of intentions (omitted). The core idea is that the non-conceptual content of perceptual mental representations can be captured in terms of perceptual similarity spaces. Perceptual concepts, i.e., the building blocks of belief-like mental representations, can be analysed semantically as regions in such a similarity space. For example, the semantics of the colour concept RED can be understood in terms of a specific region in the colour space. The action concept KICK can semantically be analysed as a region in the sensorimotor space representing bodily positions and movements (see Gärdenfors & Warglien 2012). This approach puts us in the position to satisfy the second and third criterion of adequacy of definitions of CP without conflating CP with the theory-leadenness of perception and thus violate criteria 4. Based on this account, the paper suggests the following definition of CP (here, it is only outlined).

A perceptual mental representation P R is cognitively permeated if and only if there is a cognitive mental representation C such that:

1. There is a semantic connection between the content CNPR of PR and the content CNC of C within the relevant perceptual similarity space along the lines characterised above.

2. In virtue of the causal and semantical connections between PR and C, respectively CNPR and CNC, variations in the conceptual content of C would lead to similar variations in the non-conceptual content of PR.

Subsequently, the paper shows that this definition satisfies the four criteria of adequacy for definitions of cognitive penetration.

Literature

1. Burnston, D. (2017). Cognitive penetration and the cognition-perception interface. Synthese 194, 3645–3668.

2. Firestone, C., and B. Scholl. (2015). Cognition does not affect perception: Evaluating the evidence for ’top-down’ effects. Behavioral and Brain Sciences.

3. Gärdenfors, P., Warglien, M. (2012). Using Conceptual Spaces to Model Action and Events. Journal of Semantics 29, 487–519.

4. Macpherson, F. (2012). Cognitive penetration of colour experience: Rethinking the issue in light of an indirect mechanism. Philosophy and Phenomenological Research, 84(1), 24–62.

5. Macpherson, F. (2017). The relationship between cognitive penetration and predictive coding. Consciousness and Cognition 47, 6–16.

6. Marchi F, Newen A. (2015). Cognitive penetrability and emotion recognition in human facial expressions. Frontiers of Psychology 19, doi: 10.3389/fpsyg.2015.00828. PMID: 26150796; PMCID: PMC4473593.

7. Newen A, Vetter P. (2017) Why cognitive penetration of our perceptual experience is still the most plausible account. Consciousness and Cognition 47, 26–37.

1. Pylyshyn, Z. (1999). Is vision continuous with cognition? The case for cognitive impenetrability of visual perception. The Behavioral and Brain Sciences, 22(3), 341–365; discussion 366–423.

2. Siegel, S. (2012). Cognitive penetrability and perceptual justification. Nouˆs, 46(2), 201–222.

3. Stokes, D. (2015). Towards a consequentialist understanding of cognitive penetration, eds. J. Zeimbekis, and A. Raftopoulos 2015, 75–100.

4. Toribio, J. (2015). Visual experience: Rich but impenetrable. Synthese. doi:10.1007/s11229-015-0889-8.

5. Witzel, C. (2016). An Easy Way to Show Memory Color Effects. I-Perception, 7(5). https://doi.org/10.1177/2041669516663751

The causal exclusion problem has been one of the strongest and most debated metaphysical arguments against nonreductive physicalism in the philosophy of mind over the course of the past 25 years. The core idea of the problem is that, if one is committed to a rather minimal form of nonreductive physicalism, and if some plausible (although by no means uncontroversial) principles are granted, then one is forced to admit that higher-level properties, and mental properties in particular, do not possess causal powers.

Recently, there have been several notable attempts at offering solutions to this problem based on the adoption of Woodward’s (2003) interventionist account of causation (e.g. List and Menzies (2010); Zhong (2014); Woodward (2015, 2022)), instead of directly rejecting one or more of the principles giving rise to the problem. However, it has also been argued (for example by Baumgartner (2018)) that the exclusion problem cannot be dispelled as a result of the adoption of interventionism: this would only be the case if the supervenience bases of the relevant higher-level variables were independently fixable, which they are not for metaphysical reasons.

Since the interventionist arguments against the exclusion problem differ significantly from one another, the focus of my talk will be on just one of those interventionist solutions, namely the one proposed by Woodward (2015, 2022). This particular proposal revolves around an amended version of Woodward’s original account of interventionism. The main difference between this “interventionism*” and the original version, consists in the fact that the latter is only concerned with purely causal graphs (i.e. with graphs only involving causal relations among variables), while the former explicitly takes into account causal graphs also involving metaphysical determination relations different from causation (in particular, supervenience relations). Specifically, interventionism* highlights the need not to hold the supervenience bases fixed while performing interventions on the relevant higher-level supervening variables, thus avoiding the aforementioned “independent fixability objection”.

The main claim that I will argue for in my talk is that the reasons why one may be inclined to adopt interventionism* cannot be accepted in the context of the debate over the exclusion problem. To be clear, this is not to say that one may not legitimately adopt interventionism* for pragmatic purposes (hence in conformity with the general interventionist spirit) in ordinary causal reasoning. Rather, my claim is that, when it comes to the assessment of the exclusion problem, the considerations that play a role in deciding whether some variable is to be held fixed or not tilt the scale in favour of the need to hold the variables representing supervenience bases fixed, when an intervention is performed on the variables representing the relevant supervening (mental) properties. This is because failing to do so would make it impossible to answer the causal questions that motivate the discussion around the exclusion problem itself, specifically: “Do higher-level (mental) entities possess causal powers distinct from those of their supervenience bases?”. But, as Woodward (2022, p.22) maintains, the causal questions one is interested in answering are part of the contextual factors guiding the decision to consider some variable as a confounder or not, and hence the choice of whether to hold it fixed or not. Therefore, failing to hold the supervenience bases fixed, as recommended by the amended version of interventionism, prevents one from ascertaining whether the causal powers that one may attribute to mental properties are also thereby possessed by the relevant supervenience bases, or whether they are indeed distinct.

My talk will be divided into two parts. In the first part, I will begin by offering a general presentation of the exclusion problem as a set of jointly inconsistent claims (Nonreductionism; Physicalism; Closure; Exclusion; Causal efficacy), as it has become more or less customary over time. After that, I will outline the core idea behind the interventionist account of causation: X is a direct cause of Y with respect to variable set V if there are possible interventions on X that will change the value of Y when all other variables in V are held fixed at some value by interventions (Woodward (2022, p.9). In doing so, I will explain how Woodward’s recent amended version may be used to address the exclusion problem. Then, in the second part of my talk I will argue that this revised version of interventionism should not be accepted because it does not take into account the specific theoretical interests associated with the debate over the exclusion problem, as it instead should, based on Woodward’s own recommendations.

References

Baumgartner, M. (2018). The inherent empirical underdetermination of mental causation. Australasian Journal of Philosophy, 96(2), 335-350.

Menzies, P., & List, C. (2010). The Causal Autonomy of the Special Sciences. In C. Macdonald, and G. Macdonald (eds), Emergence in Mind. Oxford: Mind Association Occasional Series.

Woodward, J. (2003). Making Things Happen: A Theory of Causal Explanation. New York: Oxford University Press.

Woodward, J. (2015). Interventionism and causal exclusion. Philosophy and Phenomenological Research, 91(2), 303-347.

Woodward, J. (2022). Modeling interventions in multi-level causal systems: supervenience, exclusion and underdetermination. European Journal for Philosophy of Science, 12(4), 59.

Zhong, L. (2014). Sophisticated exclusion and sophisticated causation. The Journal of Philosophy, 111(7), 341-360.