September 7, 2015, by Suzanne
Choice Rules and Accumulator Networks by Sudeep Bhatia
Decision makers often use sophisticated rules to make their decisions. For example, when trying to evaluate two objects they may look at a single attribute and choose the object that is best on that attribute. Alternatively, they may try to examine all attributes, and then add up these attributes to determine the total desirability of an object.
An important research question pertains to the ways in which these rules are implemented by the mind, particularly, in the types of neural networks that are responsible for processing information in the brain. Sudeep Bhatia addresses this question by showing that a well-known class of neural networks can in fact mimic the behavior generated by the sophisticated rules considered to be at play in decision making. The use of sophisticated decision rules is in fact fully compatible with simple biologically realistic information processing mechanisms.
The full paper is available here.