![Sample our Humanities journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5939/TOC-Subject-Sample-2021-Humanities.jpg"})
Abstract
The opposition between behaviour- and mind-reading accounts of data on infants and non-human primates could be less dramatic than has been thought up to now. In this paper, I argue for this thesis by analysing a possible neuro-computational explanation of early mind-reading, based on a mechanism of associative generalization which is apt to implement the notion of mental states as intervening variables proposed by Andrew Whiten. This account allows capturing important continuities between behaviour-reading and mind-reading, insofar as both are supposed to be just different kinds of generalization from perceptual experience. Specifically, I will argue that the projection of inner experiences to others which is involved in early mind-reading does not imply a computational leap beyond associative generalization from perceptual experience.
Acknowledgements
I am grateful to two anonymous referees for their very useful comments on a previous draft of this paper. I also want to thank the colleagues of the PhD on Cognitive Science at the University of Messina, and especially Pietro Perconti, for their valuable comments on an earlier presentation of this paper at their Seminars.
Notes on contributor
Marco Mazzone is an Associate Professor of Philosophy of Language at the University of Catania since 2010. His current research interests focus on neurocognition of linguistic processes, theory of intentional action and mind-reading, the nature and structure of conceptual knowledge, cognitive pragmatics.
Notes
1. In Mazzone (Citation2013), I have addressed the integration between associative processes on the one hand and attention and working memory on the other hand, with special reference to understanding communicative intentions.
2. As I have already said, in the present paper I only consider the associative dynamic subserving mind-reading, while leaving aside other components that might interact with linguistic practices as well – for instance, linguistic practices may be expected to depend on, and promote in turn, the maturation of working memory and executive functions.
3. In fact, this objection has been made by one of the referees.
4. Given the general aim of this paper, I will consider the nature of mental states in general, without any specific analysis of beliefs, desires, and other mental states.
5. From now on, I will occasionally use the phrase “mind-reading interpretation” as a shorthand for “mind-reading interpretation of the evidence on apes and/or infants”.
6. In Whiten and Suddendorf's (Citation2001, 378) terms, “[s]econdary representation adds the ability to model hypothetical situations, making possible the entertainment of multiple, simultaneous mental models”.
7. As far as I can tell, in recent studies by Baillargeon and coworkers the line of reasoning has not changed (see for instance Baillargeon et al. Citation2010).
8. In particular, I emphasize that no one, to my knowledge, has ever claimed that intervening variables imply meta-representation or explicit mastery of abstract relations between mental states: this is why intervening variables are thought to deliver only implicit mind-reading. More on this is in note 11. This is not to say that intervening variables and quasi-mental notions cannot play any role in explicit mind-reading: on the contrary, they can be meta-represented and therefore play such a role. But in this case, they become (or are on the way to becoming) full-blown mental notions.
9. In a word, Hebb's rule states that neurons that are repeatedly activated together strengthen their association.
10. Although I am only concerned here with the associative mechanism for generalization from experience, it is a reasonable hypothesis that other mechanisms (possibly including attention and working memory) contribute to explaining why quasi-mental notions are restricted to smart species (more on this below in this section).
11. To be sure, as one of the referees has observed, it could be argued that attribution of mental states implies epistemic holism (or some form of molecularism), that is, a mental state cannot be attributed unless an indefinite (or a definite) number of other mental states can be attributed too. In this case, there would be a cognitive gap after all, since mental states could not be acquired by merely generalizing over observable cues and behaviours – in other words, there would be some empirical content exceeding those cues and behaviours. However, abstract relations between mental concepts could emerge at a further stage: as a matter of fact, such relations are not part of the notion of intervening variable as formulated by Whiten (Citation1996: e.g. see the above citation) and employed in the literature discussed here. As a consequence, the possibility of epistemic holism or molecularism is not an objection to my claim that intervening variables do not have any empirical content of their own – they are just abstract rules connecting sets of observable cues and behaviours. The objection of the referee does not take seriously enough the idea that mental concepts could have a developmental course, from quasi-mental to full-blown mental states. In a sense, the objection is at risk of begging the question: if we assume that a full-blown “theory of mind” is in place from the beginning, then even quasi-mental notions can be taken to exceed observable cues and behaviours. But, as I said above, the experimental evidence I am considering here gives support to the notion of intervening variable, not to the hypothesis of an innate theory of mind. Incidentally, I propose that full-blown mental concepts emerge as a consequence of linguistic interactions: if this is true, those concepts would anyway depend on generalization over observable behaviour of some sort, specifically over linguistic behaviour.
12. It is also a conditional argument: of course, it applies in case (quasi-)mental notions are essentially conceived of as intervening variables. However, as we saw in Section 2, the notion of intervening variable is accepted from both sides of the controversy, at least within the literature we are considering here.