Vanessa Teckentrup

@glassybrain.bsky.social

897 Followers 612 Following

Neuroscience/Psychology · NeuroAdapt MSCA fellow · Digital Mental Health · Smartphone Science · Brain Stimulation · Brain-Body Communication

A graph showing higher temporal signal-to-noise ratio (tSNR) for the stimulator cable modified with floating cable traps compared to the unmodified cable.
A figure showing the association between mood and cognitive performance when the DQRT variable is lagged so we line up each assessment of cognitive processing speed with the next assessment time point for mood. Across all three datasets, there is only one item where an increase in cognitive slowing resulted in a significant change at the next assessment, here an increase in feeling restless. Hence, we concluded that there is no substantial evidence supporting the view that changes in cognitive processing speed drive subsequent changes in mood.
A figure showing the association between mood and cognitive performance when the mood variable is lagged so we line up each mood assessment with the next assessment time point for cognitive performance. Across all three datasets, feeling more worried and feeling more anxious resulted in slower response times at the next assessment. Similarly, feeling more stress resulted in cognitive slowing in two of the datasets as this item was not present in dataset 1. In turn, feeling more happy resulted in faster response times at the next assessment which we could show across the two datasets that included this positively-valenced item.
A figure showing the association between mood and cognitive performance when measured at the same time point. When we look at the within-person level, for 34 out of the 37 items across all three datasets people were slower when their negative mood increased. On the between-person level, the power usually increases due to the repeated assessments we are aggregating. In line with this, we found the same pattern with stronger effects for the between-person level, indicating that on average more negative mood was linked to slower and more positive mood (for the datasets that had both positively- and negatively-valenced items) was linked to faster response times.
A figure showing the characteristics for the 3 datasets used in the study. Dataset 1 had 753 participants who answered 16 items two times a day for 8 weeks. For responding, they used a bipolar Likert scale running from 'not at all' to 'very much so'. Dataset 2 had 173 participants who answered 11 items three times a day for 6 weeks. For responding, they used a bipolar Likert scale running from one pole of an emotion (for example 'sad') to the other pole ('happy'). Dataset 3 had 195 participants who answered 14 items two times a day for 8 weeks. For responding, they used a slider running from 'not true at all' to 'extremely true'.
A figure showing the association between mood and cognitive performance when the DQRT variable is lagged so we line up each assessment of cognitive processing speed with the next assessment time point for mood. Across all three datasets, there is only one item where an increase in cognitive slowing resulted in a significant change at the next assessment, here an increase in feeling restless. Hence, we concluded that there is no substantial evidence supporting the view that changes in cognitive processing speed drive subsequent changes in mood.
A figure showing the association between mood and cognitive performance when the mood variable is lagged so we line up each mood assessment with the next assessment time point for cognitive performance. Across all three datasets, feeling more worried and feeling more anxious resulted in slower response times at the next assessment. Similarly, feeling more stress resulted in cognitive slowing in two of the datasets as this item was not present in dataset 1. In turn, feeling more happy resulted in faster response times at the next assessment which we could show across the two datasets that included this positively-valenced item.
A figure showing the association between mood and cognitive performance when measured at the same time point. When we look at the within-person level, for 34 out of the 37 items across all three datasets people were slower when their negative mood increased. On the between-person level, the power usually increases due to the repeated assessments we are aggregating. In line with this, we found the same pattern with stronger effects for the between-person level, indicating that on average more negative mood was linked to slower and more positive mood (for the datasets that had both positively- and negatively-valenced items) was linked to faster response times.
A figure showing the characteristics for the 3 datasets used in the study. Dataset 1 had 753 participants who answered 16 items two times a day for 8 weeks. For responding, they used a bipolar Likert scale running from 'not at all' to 'very much so'. Dataset 2 had 173 participants who answered 11 items three times a day for 6 weeks. For responding, they used a bipolar Likert scale running from one pole of an emotion (for example 'sad') to the other pole ('happy'). Dataset 3 had 195 participants who answered 14 items two times a day for 8 weeks. For responding, they used a slider running from 'not true at all' to 'extremely true'.
Figure comparing fMRI recordings of key brain areas of interest for taVNS using an unmodified stimulator cable to using a cable modified with FCTs. Image artifacts are attenuated when using the modified cable.
Figure showing temperature increases across four conditions (average or maximum stimulation strength by phantom or human) measured by temperature sensors attached to the two electrodes of the stimulator during active fMRI.
On the left, the sliding mechanism of the floating ground cable traps that allows to install them on cables without opening the protective coating of the cable. On the right, the floating ground cable traps are installed on the stimulator cable and wrapped in heat-shrink tubing.
Plot showing the difference between point estimates (mean DQRT calculated from each questionnaire separately) and distributions (association between the respective questionnaire total score and mean DQRT calculated from the same number of items but (re-)drawn 100 times from any other questionnaire in the respective part of the Neureka app), indicating higher predictive performance for the total scores of the respective questionnaire for a subset of questionnaires.
Point estimates show the associations between DQRT and lifestyle, demographic, and health factors which are comparable in magnitude or exceed the predictive performance of cognitive flexibility when calculated from one survey set but are less strong for the other survey set.
Line plot showing moderate agreement for DQRT across a minimum of 15 repeated assessments within person. Two bar plots at the bottom show that both DQRT and cognitive flexibility decreased over time, indicating practice effects.
Bland-Altman plot showing a minimal deviation from zero for the difference in DQRT between subsequent assessments, indicating low bias. The plot is split into three parts for increasing time lags between the two assessments (less than 1 day, 1 to 7 days, more than 7 days). Agreement is good across all splits but decreased with increasing time lag.
Curves describing the predictive performance of mean DQRT for cognitive flexibility, working memory, and model-based planning across steps of cumulatively adding questionnaire items and recalculating mean DQRT.
Four plots showing the same pattern for point estimates reflecting the association between DQRT and cognitive flexibility, working memory, and model-based planning.
Heat map showing the highest association between DQRT and cognitive flexibility for a higher cut-off of median+2SD and no lower cut-off.
Point estimates showing that DQRT has the strongest link with cognitive flexibility, a slightly weaker but still significant link to working memory, and no link to model-based planning.