Learning rates of return is an important aspect of life for both humans and animals.  Animals must use probability learning in foraging decisions such as which patches to enter, how long to stay, when to leave.  These decisions are based on the probability of food being in each patch; that is, they based their choices on the expected rate of return.

    Humans, as well, use knowledge of rate frequently in everday life.  We must be able to choose which banks to go to, which credit cards to use, which jobs to choose, which stocks to invest in, etc.  These decisions are all made based on interest rates and rates of return.

    Many studies have focused on animals' abilities to learn different rates of reinforcement (Ferster & Skinner, 1957).  Furthermore, researchers have examined animals' abilities to choose between different rates of reinforcement.  The premise here is that if an animal can learn to respond in way that maximizes it reinforcement, it must have some sort of representation of the different rates of return. In the past, research on choice behavior typically involved the use of a T-maze (Domjan, 1998).  More recent experiments generally make use of operant chambers with concurrent schedules of reinforcement.  Operant chamber choice studies have had a main focus on concurrent VI-VI schedules, although several studies have examined both the comparison between VI and VR schedules (Reynolds, 1975) as well as concurrent VR-VR schedules (Herrnstein & Loveland, 1975; McDonald, 1988).  As well, to explore representation of rate in a more natural environment, many studies have focused on the foraging behavior of food (Godin & Keenleyside, 1984; Harper, 1982).

    Studies of VR schedules may be of particular interest as a comparison to gambling behavior in humans.  With the current controversies about Video Lottery Terminals, the learning of rates of return and response rates associated with VR schedules is an especially interesting (and applicable) topic of study.  In this paper, I will discuss theoretical models of how knowledge of rates is represented in the brain and several human examples of this representation.  Specifically, I will address how humans have a tendency to misrepresent rates of return (especially in gambling situations) by relying on heuristics and biases, and engaging in irrational thinking about chance situations.

Questions/Concerns/Links not working?   pls. contact egray@ualberta.ca