The neural development of pragmatic inference‐making in natural discourse

Abstract

Older interlocutors are more likely than younger ones to make pragmatic inferences, that is, inferences that go beyond the linguistically encoded meaning of a sentence. Here we ask whether pragmatic development is associated with increased activity in brain structures associated with inference-making or in those associated with Theory of Mind. We employed a reading task that presents vignettes in one of two versions, one of which is expected to prompt more pragmatic processing. Both versions present a major premise containing three possibilities (e.g., Xavier is considering Thursday, Friday or Saturday for inviting his girlfriend out). In the Fully-Deductive (control) condition, the major premise is followed by two disjunction-elimination premises through two separate lines (one indicating that Saturday is not convenient and another saying that Thursday is not convenient); this is followed by a valid conclusion (e.g., "I'll reserve Friday"). In the Implicated-Premise condition, the first disjunction-elimination premise is followed by a second similar one that eliminates the same disjunction (e.g., both lines explain why Saturday is not convenient). In this condition, readers may pragmatically enrich the conclusion (i.e., "I'll reserve Friday" pragmatically implicates that Xavier is also eliminating Thursday from consideration). Reading times in Experiment 1 showed that processing the speaker's conclusion in the Implicated-Premise condition becomes increasingly more effort-demanding as readers reach adolescence. Experiment 2 showed that this developmental pattern is related to age-related increases in fMRI activity in fronto-parietal regions typically involved in inference-making processes. We found no evidence indicating age effects related to Theory of Mind areas.

Dissociating Language and Thought in Human Reasoning

“…Although we tested the language-centric hypothesis of deduction in the context of a specific mode of deductive reasoning (i.e., propositional logic), previous work suggests that this conclusion may be expected to also apply to categorical syllogisms [54,73,115], relational problems [116], and pragmatic inferences in the context of naturalistic discourse [117,118]. It should be noted, however, that while our findings imply that the mechanisms of natural language do not participate in the syntax-like operations of deduction, they do not rule out the Vygotskyan idea that language may serve, throughout development, as a "cognitive scaffolding" [9] enabling the acquisition of structure-dependent operations in other domains, such as logic, to then become independent in adulthood.…”

Dissociating language and thought in human reasoning

“…Although we tested the language-centric hypothesis of deduction in the context of a specific mode of deductive reasoning (i.e., propositional logic), previous work suggests that this conclusion can reasonably be expected to extend to categorical syllogisms (Rodriguez-Moreno and Hirsch 2009;Tsujii et al 2011;Prado et al 2013), relational problems (Knauff et al 2003), and pragmatic inferences in the context of naturalistic discourse (Prado et al 2015;Schwartz et al 2018). The findings from this study, however, are still compatible with the Vygotskyan idea that language may serve, throughout development, as a "cognitive scaffolding" (Carruthers 2002) enabling the acquisition of structure-dependent operations such as those of logic, to then become independent, in adulthood.…”

“…Although we tested the language-centric hypothesis of deduction in the context of a specific mode of deductive reasoning (i.e., propositional logic), previous work suggests that this conclusion may be expected to also apply to categorical syllogisms [37,52,97], relational problems [98], and pragmatic inferences in the context of naturalistic discourse [99,100]. It should be noted, however, that while our findings imply that the mechanisms of natural language do not participate in the syntax-like operations of deduction, they do not rule out the Vygotskyan idea that language may serve, throughout development, as a “cognitive scaffolding” [101] enabling the acquisition of structure-dependent operations in other domains, such as logic, to then become independent in adulthood.…”

“…This result is consistent with neuropsychological evidence demonstrating that patients 129 with lesions spanning frontomedial cortices (including our cTBS site in dorsomesial 130 frontal cortex, BA8) are impaired at deductive reasoning despite no observable 131 structural damage in Broca's area and ceiling performance on standard 132 neuropsychological tests of language [36]. 133 Although we tested the language-centric hypothesis of deduction in the context of a 134 specific mode of deductive reasoning (i.e., propositional logic), previous work suggests 135 that this conclusion can reasonably be expected to extend to categorical 136 syllogisms [27,37,38], relational problems [39], and pragmatic inferences in the context 137 of naturalistic discourse [40,41]. The findings from this study, however, are still 138 compatible with the Vygotskyan idea that language may serve, throughout development, 139 as a "cognitive scaffolding" [3] enabling the acquisition of structure-dependent 140 operations such as those of logic, to then become independent, in adulthood.…”

The neural bases of argumentative reasoning

“…In additional analyses, we extracted brain activity from regions of interest (ROIs) that were identified in a manual coordinate-based meta-analysis of the ToM network (Van Overwalle & Baetens, 2009). These ROIs, which were also used in previous studies from our group (Schwartz, Epinat-Duclos, Noveck, & Prado, 2018;Spotorno et al, 2012)…”