Decision Weights for Experimental Asset Prices Based on Visual Salience

Abstract

We apply a machine-learning algorithm, calibrated using general human vision, to predict the visual salience of prices of stock price charts. We hypothesize that the visual salience of adjacent prices increases the decision weights on returns computed from those prices. We analyze the inferred impact of these weights in two experimental studies that use either historical price charts or simpler artificial sequences. We find that decision weights derived from visual salience are associated with experimental investments. The predictability is not subsumed by statistical features and goes beyond established models.