One of the most useful tools in the service of communication is language. However, the neuro-cognitive basis of language has often been studied outside of its natural niche and in isolation from its communicative function. This thesis examines the neuro-cognitive processes that are at the basis of processing communicative intention conveyed by language by employing a range of psycho- and neurolinguistic methods. In particular, the present work focuses on two pragmatic phenomena: speech acts and indirect speech acts. The following questions are asked: (1) Can the differences in neural signatures of speech acts previously observed in the comprehension modality also be found in speech production? (2) Is the right temporo-parietal junction, an important node of the ToM network, causally involved in the comprehension of indirect speech acts? (3) Do indirect speech acts systematically differ from direct ones in psycholinguistic properties, whose processing is known to be reflected in different neural processes? The general methodological approach taken here is to use identical words or sentences but alter their pragmatic-communicative roles by embedding them in different dialogic or situational contexts. This way, the communicative function can be examined independently from the linguistic form used to carry it out.
In a first study, the neural representations of naming and request actions were examined. These were performed using the same utterance during speech production, while subjects participated in an interactive communicative task and while participants neural activity was recorded by electroencephalography. The aim was to compare these findings to previous findings in the comprehension modalities. We find that uttering the same words with different speech act functions (naming and request) is associated with different electrophysiological signatures. These differences are similar to those found when comparing the same two speech acts in the comprehension modality. In particular, requests are associated with activations of the motor system, supporting the idea that their intrinsic link to action is also encoded in the brain.
The second study tested whether the comprehension of indirect speech acts relies on the right-temporoparietal junction, a brain region thought to contribute to Theory of Mind processes. To do so, activity in these brain regions was altered by means of (non-invasive) transcranial magnetic stimulation. Subjects were then exposed to indirect speech acts and their matched direct controls, and their comprehension processes were behaviorally monitored. The finding that comprehending indirect speech acts is more costly than comprehending direct ones was replicated. Applying TMS to the right-temporoparietal junction did not affect the processing of indirect speech acts when these were matched to their direct controls in terms of communicative function. However, the speed of comprehension of indirect speech acts was altered relative to the direct controls when they were not matched for communicative function.
In a third study, subjects were asked to provide ratings of several psycholinguistic dimensions for both direct and indirect speech acts to assess the differences between them. Compared to their direct counterparts, indirect speech acts were found to be less predictable, less coherent with their context, less semantically related to their context, and understood with less certainty. Notably, these properties were tightly related to the in/directness of the stimuli.
In summary, it could be shown that (i) communicative function can be encoded in the brain in ways that are similar between comprehension and production modality, (ii) specificities of the neural representations of speech acts can be related to their use in communication, (iii) there was no evidence of the right-temporoparietal junction processing indirect speech acts when compared to well-matched controls, and (iv) contrasting direct and indirect speech acts revealed several differences unrelated to ToM that suggest the (additional) contribution of other brain systems in the comprehension of indirect speech acts. Overall, it was demonstrated that when identical utterances are used with different communicative functions—whether direct or indirect—are associated with different neurocognitive processes. These findings add to the growing literature examining the communicative function of language and argue for greater inclusion of pragmatics in neurocognitive models of language function.
