The Temporal Neural Dynamics of Math and Reading in the Adult Human Brain

956 

Abstract Submission 

Gizem Cetin^1,2, Mareike Grotheer^1,2, Daniel Kaiser^3,2

¹Department of Psychology, Philipps University Marburg, Marburg, Germany, ²Center for Mind, Brain, and Behavior (CMBB), Marburg, Germany, ³Mathematical Institute, Justus-Liebig University Gießen, Gießen, Germany

Gizem Cetin  
Department of Psychology, Philipps University Marburg|Center for Mind, Brain, and Behavior (CMBB)
Marburg, Germany|Marburg, Germany

Mareike Grotheer  
Department of Psychology, Philipps University Marburg|Center for Mind, Brain, and Behavior (CMBB)
Marburg, Germany|Marburg, Germany

Daniel Kaiser  
Mathematical Institute, Justus-Liebig University Gießen|Center for Mind, Brain, and Behavior (CMBB)
Gießen, Germany|Marburg, Germany

Math and reading abilities play a fundamental role in our everyday life, influencing the well-being and career prospects of individuals. Neuroimaging studies have revealed that these skills rely on specialized brain networks spanning from the ventral temporal cortex to the frontal cortex (e.g., Grotheer et al., NComm, 2019). However, the precise temporal dynamics of information exchange within these networks remain unclear. To gain insights into the temporal dynamics of mathematical and reading processing in the brain, we evaluated neural responses during math and reading tasks in the EEG, using an established experimental design where both tasks are performed on the same visual stimuli.

At the onset of each trial, a cue was presented. Participants (N=30) then viewed four number-letter morph stimuli, each for 1 second, followed by a 2-second answer screen, during which they provided responses through button presses. Across different runs, participants performed three different tasks: reading, math, and a control task (color discrimination) (Fig. 1). Multivariate pattern analyses were conducted on EEG sensor data to evaluate the spatio-temporal dynamics of information flow during the tasks. Specifically, we tested when different groups of electrodes carry information about the participants' task, whether participants viewed more number- or letter-like morphs, and the resulting word or sum.

We found that the task participants were performing could first be decoded from electrodes in the frontal cortex and only later be decoded from electrodes in the occipito-temporal cortex. These temporal dynamics were similar across hemispheres, but decoding accuracy was overall higher in the left hemisphere. We could not decode the specific morph stimulus that the participants were looking at. Similarly, we could not significantly decode the resulting word/sum in either the math or the reading task.

We were able to decode the task participants were performing but not the visual stimulus they were seeing, which suggests that the performed task has a greater impact on neural responses than the visual stimuli. Further, our results suggest that information about current task demands can first be found in the frontal cortex and is subsequently fed back into the visual system to modulate processing in a task-specific way. This finding suggests that frontal control systems play a critical role in formatting visual representations of the same visual inputs for efficient use during math and reading tasks. Future research combining EEG and fMRI could resolve neural information flows during these essential tasks with even greater spatial precision.

Higher Cognitive Functions Other ¹

Reading and Writing

EEG/MEG Modeling and Analysis

EEG ²

Perception: Visual

Cognition

Electroencephaolography (EEG)

Language

Perception

Vision