Valence Asymmetry in Cognition - A Formal Account

Most psychological theories are expressed in natural language terms, which incurs a number of important problems (e.g., vagueness, limited refutability, lack of parsimony, possible contradictiveness, redundancy and specification gaps). Better theory specification has been called for many times, and formalization is a means of achieving it. We present a formal account of “valence asymmetry”, that is: the notion that positive and negative stimuli affect information processing differentially. Most often, negative stimuli are assumed to have a stronger impact (“negativity bias”). We formalize two of the major frameworks that have been offered as explanations for this alleged asymmetry: The first framework emphasizes the role of valence as a trigger for deeper information processing. It posits that negative stimuli are more survival-relevant than positive ones. The second framework emphasizes the relative similarity and frequency of stimuli. Here, it is assumed that negative stimuli are less similar to one another and less frequent, and therefore afford greater processing effort. Valence itself does not affect information processing within this framework. Our formal analysis highlights the conceptual complexity of these issues, which remains hidden when using the traditional narrative form. For example, we explicate several implicit assumptions that are necessary to arrive at some of the major predictions about valence asymmetry. Both frameworks may be expressed using a single set of parameters, with considerable (but not perfect) overlap. Formalizing valence asymmetry explanations is clearly advantageous in terms of exactness, parsimony, and comparative testability.