Cortical travelling waves relate to variation in personality traits

Abstract

Personality traits must relate to stable neural processes, yet few robust neural correlates of personality have been discovered. Recent methodological advances enable measurement of cortical travelling waves, which likely underpin information flow between brain regions. Here, we explore whether cortical travelling waves relate to personality traits from the "Big Five" taxonomy. We assessed personality traits and recorded resting electroencephalography (EEG) from 300 participants. We computed travelling wave strength using a 3D fast Fourier transform and explored relationships between alpha travelling waves and personality traits. Trait Agreeableness and Openness/Intellect had significant relationships to travelling waves that passed multiple-comparison controls (p _FDR = 0.019 and p _FDR = 0.036 respectively). Agreeableness related to interhemispheric waves travelling from the right hemisphere along central lines (rho = 0.263, p < 0.001, BF10 = 356.350). This relationship was unique to the compassion aspect (t = 3.719, p < 0.001) rather than politeness aspect of Agreeableness (t = 0.897, p = 0.370). Openness/Intellect related to backwards travelling waves along midline electrodes (rho = 0.197, p < 0.001, BF10 = 13.800), which was confirmed for the Openness aspect (rho = 0.216, p < 0.001, BF10 = 26.444) but not the Intellect aspect (rho = 0.093, p = 0.109, BF10 = 0.344). Greater cortical travelling wave strength from right temporal regions was associated with higher trait compassion, and backwards travelling wave strength along midline electrodes was associated with trait openness. Further research is needed to investigate the mechanistic role of travelling waves in personality traits and other individual differences.

Keywords: compassion; cortical travelling waves; electroencephalography; openness; personality traits.

Fig. 1.

A visual representation of the travelling wave computation method using a 2-dimensional fast Fourier transform (2D-FFT), which can be extrapolated to understand the 3-dimensional fast Fourier transform (3D-FFT) used in our primary analyses. (A, B) First, 1 s epochs were extracted from the continuous electroencephalography (EEG) data after removal of artifacts. The time-series from each electrode were then organised into matrices, with posterior electrodes at the bottom of the matrices and anterior electrodes at the top (and left electrodes on the left side of the 3D matrices, and right electrodes on the right side), as depicted in (A, B) (after applying a bandpass filter in the alpha frequencies of interest to highlight the travelling waves). (C) Each epoch was then used as the input for a 3D-FFT (MATLAB), or 2D-FFT in the case of our post-hoc tests (as depicted above for simplification of the visualisation). This provided a matrix providing power values in each spatial and temporal frequency. Within this matrix, the x-axis provides power values at each temporal frequency, while the y-axis provides power values at each spatial frequency—shown in C with values above 0 reflecting backwards travelling spatial waves (from frontal to posterior electrodes), values below 0 reflecting forwards travelling spatial waves (from posterior to frontal electrodes), and values at 0 reflecting standing waves (waves that do not travel). In our 3D-FFT (not depicted in this figure), the z-axis would reflect lateral travelling waves (from left to right or right to left). (D) Next, to assess the travelling wave strength in each spatial and temporal frequency, we divided the power in each cell in the matrix by the power obtained after randomly shuffling the order of the electrodes in each epoch 100 times, repeating the FFT computations for each shuffle, and averaging the outputs of these null FFTs. We then computed a log10 transform on the output of this division and multiplied this value by 10 to obtain a measure of the strength by which the real travelling wave strength exceeded the permutation-based null travelling wave strength, on the decibel scale (D).

Fig. 2.

Spearman’s correlations between 3D-FFT travelling wave values at each spatial frequency and direction, and at 10 Hz in the frequency domain, and agreeableness (top) or openness/intellect (bottom). * indicates cells that were involved within a cluster that passed our initial cluster statistical test of relationships between the personality traits and 3D-FFT outcomes (p_FDR < 0.05), with values in each cell representing travelling waves across the directions that could be measured across the scalp electrodes.

Fig. 3.

Scatterplots showing the relationship between agreeableness (top left), compassion (top right), or politeness (bottom left) and right to left cortical travelling alpha waves (at 9 to 10 Hz, computed using a 2D-FFT from the central electrode line, units in decibels) within the eyes open resting data. The blue lines indicate the linear regression line with the shaded area indicating the 95% confidence interval for this regression line.

Fig. 4.

Raincloud plots depicting participants from the top and bottom 10^th percentile of right to left travelling wave strengths (9–10 Hz) and their associated scores for compassion (top) and agreeableness (bottom).

Fig. 5.

Scatterplots showing the relationship between Openness/Intellect (top left), openness (top right), or intellect (bottom left) and backwards travelling alpha waves (at 9 to 10 Hz, computed using a 2D-FFT from the midline electrodes, units in decibels) within the eyes open resting data. The blue lines indicate the linear regression line with the shaded area indicating the 95% confidence interval for this regression line.

Fig. 6.

Raincloud plots depicting participants from the top and bottom 10^th percentile of backwards travelling wave strengths (10–11 Hz) and their associated scores for Openness (top) and Openness/Intellect (bottom).