Left Frontal EEG Power Responds to Stock Price Changes in a Simulated Asset Bubble Market.

Filip-Mihai Toma,Makoto Miyakoshi

doi:10.3390/brainsci11060670

Abstract

Financial bubbles are a result of aggregate irrational behavior and cannot be explained by standard economic pricing theory. Research in neuroeconomics can improve our understanding of their causes. We conducted an experiment in which 28 healthy subjects traded in a simulated market bubble, while scalp EEG was recorded using a low-cost, BCI-friendly desktop device with 14 electrodes. Independent component (IC) analysis was performed to decompose brain signals and the obtained scalp topography was used to cluster the ICs. We computed single-trial time-frequency power relative to the onset of stock price display and estimated the correlation between EEG power and stock price across trials using a general linear model. We found that delta band (1–4 Hz) EEG power within the left frontal region negatively correlated with the trial-by-trial stock prices including the financial bubble. We interpreted the result as stimulus-preceding negativity (SPN) occurring as a dis-inhibition of the resting state network. We conclude that the combination between the desktop-BCI-friendly EEG, the simulated financial bubble and advanced signal processing and statistical approaches could successfully identify the neural correlate of the financial bubble. We add to the neuroeconomics literature a complementary EEG neurometric as a bubble predictor, which can further be explored in future decision-making experiments.

Highlights

Crashes of financial bubbles have had severe negative impacts on society
Based on our hypothesis that the cognitive processes involved in financial decision-making should be related to the lateral frontal functions, we examined this left frontal Independent component (IC) cluster for further analysis
The goal of the current study is to test whether the trial-by-trial EEG signal change corresponding to stock price dynamics in a financial bubble can be detected in the lateral frontal regions, as predicted in the neuroimaging literature

Summary

Introduction

Bubbles are deviations from the market equilibrium for which standard asset pricing theory does not apply, and they form when prices are significantly increased above a fundamental intrinsic value [1,2] Their causes could be attributed to over-inflated expectations of market participants towards future prices, which in turn are a consequence of cognitive biases such as aggregate overconfidence [1] or herding behavior [3]. These psychological factors, what Keynes deemed as “animal spirits”, drive deviations from market equilibrium, warranting supplementary methods from the field of experimental cognitive neuroscience to understand bubble dynamics in addition to the traditional econometric approaches using historical data [4]. The use of low-cost desktop/wearable EEG-based BCI devices may become a standard practice in the near future for stock trading to support the user’s decisions or to calibrate behavior and provide real time neurofeedback in this sense [5]

Objectives

Methods

Results

Discussion

Conclusion