E-field Modelling Explains Variability in Behavioural Response to Chronometric TMS During N-back
Poster No:
110
Submission Type:
Abstract Submission
Authors:
Sarah Grosshagauer1, Maria Vasileiadi1, Anna-Lisa Schuler2, Michael Woletz1, Christian Windischberger3, Martin Tik4
Institutions:
1Medical University of Vienna, Vienna, Vienna, 2Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Saxony, 3Medical University of Vienna, Vienna, Austria, 4High Field MR Center, Center for Medical Physics and Biomedical Engineering, Medical University of V, Vienna, Austria
First Author:
Co-Author(s):
Martin Tik
High Field MR Center, Center for Medical Physics and Biomedical Engineering, Medical University of V
Vienna, Austria
High Field MR Center, Center for Medical Physics and Biomedical Engineering, Medical University of V
Vienna, Austria
Introduction:
Transcranial Magnetic Stimulation (TMS) allows for non-invasive, transient modulation of human brain activity, affecting cognitive functions. Chronometric TMS offers a time-sensitive window into studying how brain regions contribute to behaviour during different phases of task processing. However, TMS effects are associated with high inter-individual variation (Hartwigsen et al. 2022). We performed chronometric TMS-fMRI by stimulating an individualised left dorsolateral prefrontal cortex (DLPFC) target during precisely defined timepoints of N-back task processing. While the primary focus of this study was on modulation of clinically relevant network nodes, high subject-to-subject variability in behavioural changes was identified. Thus, we aimed to evaluate whether variation in induced electric field (E-field) could explain differences in behavioural response.
Methods:
We performed chronometric TMS during an N-back task [2] in 14 healthy participants (9 female/5 male, age mean±std 24±5.8 y). TMS was applied to the portion of left DLPFC most anti-correlated to subgenual anterior cingulate cortex (sgACC) [3]. Participants received 10 Hz triplets of TMS at 100% of resting motor threshold at two precisely defined timepoints (pre or post letter onset) of the task paradigm (figure 1a). For both cognitive loads (0- and 2–back), stimulation was performed on target letters as well as randomly selected non-target letters. Target letters were defined as the second letter of the 2-back pair (decoding). Additionally, participants were invited for a second measurement, where the chronometric stimulation protocol was modified: during 2-back TMS was exclusively applied during random letters and first letters of the 2-back pairs (encoding). Based on individual T1 scans and the recorded TMS coil position and orientation, E-field models were created for each participant using SimNIBS 4.0.0 [4]. E-fields were transformed to MNI space and smoothed (FWHM=6 mm) to reduce effects of individual gyrification. To assess behaviour, we evaluated reaction time (RT) for two task difficulties and each timing condition separately. To account for individual variability in subject-specific reaction time, RT in TMS-blocks was divided by the average RT of the same task difficulty without TMS, resulting in a reaction time ratio (RTR). RTRs of each task/timing combination were used for voxel-wise correlation between electric fields and the behavioural covariates using SPM12. Correlation was performed within areas with mean E-field ≥ 30 V/m. Thresholds for statistical significance were defined at p<0.05.
Results:
RTRs were statistically significantly different to one on a group level for 0-back stimulation prior to letters, i.e. TMS induced a statistically significant (p<0.001) group level decrease in reaction time compared to baseline (RTR<1, figure 2a). In addition, 0-back TMS post letter (p=0.025) and 2-back stimulation prior to letter appearance (encoding) were statistically significant at p<0.05. Maps showing correlation coefficients between E-field and RTR are depicted in figure 2. For the 0-back task both timings resulted in similar patterns, however, significantly different correlated clusters were identified. Interestingly, high E-fields close to the 2-back activation hotspot (peak: -40, 26, 36 [MNI]; R=0.76; p<0.001) showed positive correlation with RTR if stimulation was performed at the pre-timepoint, especially if stimulation occurred on target letters.
Conclusions:
Correlation analysis between RTR and E-fields shows that induced E-fields contribute to inter-individual variability in behaviour. Specific timings of the same task condition resulted in different correlation maps. This indicates that TMS at different timepoints result in either enhancements or disruptions of the ongoing and subsequent neural processing. In conclusion, these results indicate distinct prefrontal subregions contributing to early vs. late phases of N-back task processing.
Brain Stimulation:
TMS 1
Learning and Memory:
Working Memory 2
Modeling and Analysis Methods:
Other Methods
Keywords:
Cognition
Data analysis
Memory
Statistical Methods
STRUCTURAL MRI
Transcranial Magnetic Stimulation (TMS)
Other - Electric Field Modelling
1|2Indicates the priority used for review
Provide references using author date format
Hartwigsen, Gesa (2022). “Noninvasive Brain Stimulation: Multiple Effects on Cognition.” The Neuroscientist, July, 10738584221113806. https://doi.org/10.1177/10738584221113806.
Owen, Adrian M. (2005). “N‐back Working Memory Paradigm: A Meta‐analysis of Normative Functional Neuroimaging Studies.” Human Brain Mapping 25 (1): 46–59. https://doi.org/10.1002/hbm.20131.
Thielscher, Axel (2015). “Field Modeling for Transcranial Magnetic Stimulation: A Useful Tool to Understand the Physiological Effects of TMS?” In 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), 222–25. https://doi.org/10.1109/EMBC.2015.7318340.