This article examines the idea of mind-reading technology by focusing on an interesting case of applying a large language model (LLM) to brain data. On the face of it, experimental results appear to show that it is possible to reconstruct mental contents directly from brain data by processing via a chatGPT-like LLM. However, the author argues that this apparent conclusion is not warranted. Through examining how LLMs work, it is shown that they are importantly different from natural language. The former (...) operates on the basis of nonrational data transformations based on a large textual corpus. The latter has a rational dimension, being based on reasons. Using this as a basis, it is argued that brain data does not directly reveal mental content, but can be processed to ground predictions indirectly about mental content. The author concludes that this is impressive but different in principle from technology-mediated mind reading. The applications of LLM-based brain data processing are nevertheless promising for speech rehabilitation or novel communication methods. (shrink)

Philosophers and psychologists assume that mental imagery is a cognitive state, that it represents things as being a certain way. However, I argue that imagery is a conative state: it represents things as to be made a certain way. I challenge the traditional assumption by targeting an increasingly popular cognitive account that identifies mental imagery, such as inner speech, with predictions of sensory input. This predictive account faces both empirical and theoretical problems. The account not only fails to capture the (...) salience effects exhibited by mental imagery, it also cannot explain why subjects should not spend their lives minimizing prediction error by engaging in fantasy. These shortcomings highlight what I call the conative profile of imagery, which includes its salience, motivational, and performance effects. I argue that an alternative, conative view of imagery, according to which mental images are goal states, is best able to account for its conative profile. (shrink)

Theorists have found it difficult to reconcile the unity of inner speech as a mental state kind with the diversity of its manifestations. I argue that existing views concerning the content of inner speech fail to accommodate both of these features because they mistakenly assume that its content is to be found in the ‘speech processing hierarchy’, which includes semantic, syntactic, phonemic, phonetic, and articulatory levels. Upon rejecting this assumption, I offer a position on which the content of inner speech (...) is determined by voice processing, of which speech processing is but one component. The resulting view does justice to the idea that inner speech is a motley assortment of episodes that nevertheless form a kind. (shrink)

Covert speech is accompanied by a subjective multisensory experience with auditory and kinaesthetic components. An influential hypothesis states that these sensory percepts result from a simulation of the corresponding motor action that relies on the same internal models recruited for the control of overt speech. This simulationist view raises the question of how it is possible to imagine speech without executing it. In this perspective, we discuss the possible role played by motor inhibition during covert speech production. We suggest that (...) considering covert speech as an inhibited form of overt speech maps naturally to the purported progressive internalization of overt speech during childhood. We further argue that the role of motor inhibition may differ widely across different forms of covert speech and that considering this variety helps reconciling seemingly contradictory findings from the neuroimaging literature. (shrink)

In updating our beliefs on the basis of our background attitudes and evidence we frequently employ objects in our environment to represent pertinent information. For example, we may write our premises and lemmas on a whiteboard to aid in a proof or move the beads of an abacus to assist in a calculation. In both cases, we generate extramental (that is, occurring outside of the mind) representational states, and, at least in the case of the abacus, we operate over these (...) states in light of their contents (e.g., the integers represented by the beads) to generate new representations. In this paper, I argue that our belief updating processes and the grounds of their rational evaluation are partly constituted by extramental representations and operations. In other words, we don’t merely update our attitudes through an internal process of reasoning on the basis of available evidence. If we are to accurately understand and rationally evaluate our belief updating processes and resultant attitudes, we need to examine how we representationally appropriate our extramental environment in the updating process. (shrink)

Our conscious thought, at least at times, seems suffused with language. We may experience thinking as if we were “talking in our head”, thus using inner speech to verbalize, e.g., our premises, lemmas, and conclusions. I take inner speech to be part of a larger phenomenon I call inner semiotics, where inner semiotics involves the subjective experience of expressions in a semiotic (or symbol) system absent the overt articulation of the expressions. In this paper, I argue that inner semiotics allows (...) us to bootstrap our way into entertaining thoughts about exact numbers and quantities that we couldn't prior to our competence with a numeric code. I establish that our arithmetic thoughts literally occur as (internal ‘articulations’ of) expressions in a numeric code. However, a problem arises for my view: just as we can slip in overt speech, producing an utterance that deviates from what we mean to say, there is very good evidence that we can slip in inner speech as well. If our arithmetic thought occurs in a numeric code, it's far from clear how we determine when a covert utterance constitutes a slip. In closing, I provide an account of what makes an inner speech utterance a slip. (shrink)

While the idea that thinking is a form of silent self-talk goes back at least to Plato, it is not immediately clear how to state this thesis precisely. The aim of the paper is to spell out the notion that we think in language by recourse to recent work on inner speech. To that end, inner speech and overt speech are briefly compared. I then propose that inner speaking be defined as a mental episode that substantially engages the speech production (...) system; the underlying model of speech production is sketched. Next, the cognitive role of inner speaking is explored, especially its role in thinking and reasoning. The question of whether it is a way of making thoughts accessible (to whomever) or whether it is a means of thinking itself is raised. I argue that there are two reasons for assuming that, occasionally, we think in language. More specifically, I will claim that some instances of thinking are instances of inner speaking as they exploit certain properties of natural language, and that some instances of inner speaking are instances of thinking as they play a decisive role in paradigmatic cases of thinking that result from internalizing and re-using certain social-linguistic practices. Finally, the Language-of-Thought hypothesis as an alternative account is critically discussed. (shrink)

Predictions during language comprehension are currently discussed from many points of view. One area where predictive processing may play a particular role concerns poetic language that is regularized by meter and rhyme, thus allowing strong predictions regarding the timing and stress of individual syllables. While there is growing evidence that these prosodic regularities influence language processing, less is known about the potential influence of prosodic preferences on neurophysiological processes. To this end, the present electroencephalogram study examined whether the predictability of (...) strong and weak syllables within metered speech would differ as a function of meter. Strong, i.e., accented positions within a foot should be more predictable than weak, i.e., unaccented positions. Our focus was on disyllabic pseudowords that solely differed between trochaic and iambic structure, with trochees providing the preferred foot in German. Methodologically, we focused on the omission Mismatch Negativity that is elicited when an anticipated auditory stimulus is omitted. The resulting electrophysiological brain response is particularly interesting because its elicitation does not depend on a physical stimulus. Omissions in deviant position of a passive oddball paradigm occurred at either first- or second-syllable position of the aforementioned pseudowords, resulting in a 2-by-2 design with the factors foot type and omission position. Analyses focused on the mean oMMN amplitude and latency differences across the four conditions. The result pattern was characterized by an interaction of the effects of foot type and omission position for both amplitudes and latencies. In first position, omissions resulted in larger and earlier oMMNs for trochees than for iambs. In second position, omissions resulted in larger oMMNs for iambs than for trochees, but the oMMN latency did not differ. The results suggest that omissions, particularly in initial position, are modulated by a trochaic preference in German. The preferred strong-weak pattern may have strengthened the prosodic prediction, especially for matching, trochaic stimuli, such that the violation of this prediction led to an earlier and stronger prediction error. Altogether, predictive processing seems to play a particular role in metered speech, especially if the meter is based on the preferred foot type. (shrink)