Neurological Perspectives on Pragmatic Inference: Examining Conversational Sentences

1008 

Abstract Submission 

Jiseon Baik¹, Sole Yoo^2,3, Euisun Kim^4,3, Jiyoung Park^2,3, Jiho Min¹, Hae-Jeong Park^2,3,4,5, Haeil Park¹

¹Kyung Hee University, Seoul, Korea, Republic of, ²Department of Cognitive Science, Yonsei University, Seoul, Korea, Republic of, ³Department of Nuclear Medicine, Severance Hospital, Yonsei University College of Medicine, Seoul, Korea, Republic of, ⁴Department of Nuclear Medicine, Graduate School of Medical Science, BK 21 Project, Yonsei University, Seoul, Korea, Republic of, ⁵Center for Systems and Translational Brain Sciences, Institute of Human Complexity and Systems Science, Yonsei University,, Seoul, Korea, Republic of

Jiseon Baik  
Kyung Hee University
Seoul, Korea, Republic of

Sole Yoo  
Department of Cognitive Science, Yonsei University|Department of Nuclear Medicine, Severance Hospital, Yonsei University College of Medicine
Seoul, Korea, Republic of|Seoul, Korea, Republic of

Euisun Kim  
Department of Nuclear Medicine, Graduate School of Medical Science, BK 21 Project, Yonsei University|Department of Nuclear Medicine, Severance Hospital, Yonsei University College of Medicine
Seoul, Korea, Republic of|Seoul, Korea, Republic of

Jiyoung Park  
Department of Cognitive Science, Yonsei University|Department of Nuclear Medicine, Severance Hospital, Yonsei University College of Medicine
Seoul, Korea, Republic of|Seoul, Korea, Republic of

Jiho Min  
Kyung Hee University
Seoul, Korea, Republic of

Hae-Jeong Park  
Department of Cognitive Science, Yonsei University|Department of Nuclear Medicine, Severance Hospital, Yonsei University College of Medicine|Department of Nuclear Medicine, Graduate School of Medical Science, BK 21 Project, Yonsei University|Center for Systems and Translational Brain Sciences, Institute of Human Complexity and Systems Science, Yonsei University,
Seoul, Korea, Republic of|Seoul, Korea, Republic of|Seoul, Korea, Republic of|Seoul, Korea, Republic of

Haeil Park  
Kyung Hee University
Seoul, Korea, Republic of

With the rise of social media, people have increasingly used emojis and small digital images, often featuring a face, as a means of communication. However, because of its nature, emojis can lack verbal meaning, which sometimes leads to confusion in our communication. Prior to the emergence of emoji, people used interjections such as "wow!" and "gross!" to express emotions, which are context-specific linguistic signs. By understanding the impact of these elements on language use and comprehension, we can enhance our understanding of the role of verbal cues in communication. In some ways, both expressions play a similar role in everyday conversations, and therefore, we aimed to investigate whether different brain regions are involved depending on the form of expression.

Thirty-three right-handed native English speakers (18 women) with normal vision and no neurological disorders participated in this imaging study. All participants provided written informed consent prior to participation. The study involved two sessions, each with 16 sentences, divided into four conditions: Type, Congruency, and Session. A total of 64 English sentence stimuli were used in this study. During the experiment, participants were asked to make judgments about each sentence, and the session lasted for approximately six minutes. The participants were scanned using a 3-T MRI system (Philips 3-T magnet) and functional imaging was performed to measure brain activation via the BOLD signal. For group-level inferences, a cluster-level criterion with a voxel-level threshold of P < 0.005 and a cluster size of >196 was used.

The imaging experiment revealed significant differences in brain activation patterns between the various conditions. The contrast between 'Emoji' and 'Interjection' conditions demonstrated greater activation in the fusiform gyrus and right inferior frontal gyrus for 'Emoji,' while 'Interjection' showed greater activation in the left superior temporal gyrus (STG), right STG, middle temporal gyrus (MTG). The 'Incongruent' condition relative to 'Congruent' condition demonstrated greater activity in the regions associated with error detection, including the anterior cingulate cortex (ACC) and right inferior frontal gyrus (IFG). Additionally, the contrast between 'Session 2' and 'Session 1' indicated increased activation in the supramarginal gyrus, angular gyrus, and frontal operculum for 'Session 2'.

The study found that activation in language-related areas and the superior temporal gyrus is associated with speech perception. Despite participants reading interjections instead of hearing them, activation in the speech perception areas was anticipated. In the emoji condition, activation in the fusiform gyrus (facial expressions) and right inferior frontal gyrus suggested that participants connected sentence meaning with emojis, potentially involving facial imagery. This aligns with the role of the right inferior frontal gyrus in syntactic reanalysis and linguistic judgement (Mack et al., 2013).

In incongruent situations, participants tended to undergo reanalysis, whether the sentence ended with an emoji or an interjection. Increased complexity in mismatches prompts more inferences and context reinterpretation. Inhibitory control is crucial for resolving conflicts and suppressing automatic responses triggered by incongruent stimuli in mismatch tasks. Notably, 'Session 2' vs. 'Session 1' showed increased activation in the supramarginal gyrus, associated with social context, suggesting participants speculated about others' mental states during the experiment.

Language Comprehension and Semantics ¹

Activation (eg. BOLD task-fMRI) ²

BOLD fMRI

Data analysis

FUNCTIONAL MRI

Language