|
SOCR MODEL: Stout, S.C. & Miller, R.R. (in press) On this page is sample simulation code for the SOCR model. SOCR stands for SOmetimes Competing Retrieval and is a formalization and expansion of the extended comparator hypothesis of Denniston, Savastano, & Miller (2001). The article is coauthored with Ralph Miller of Binghamton University. A description of the model is published in Psychological Review. The full reference is: Stout, S. C., & Miller, R. R. (2007). Sometimes competing retrieval (SOCR): A formalization of the comparator hypothesis. Psychological Review, 114, 759-783. Simulations of the model used in this manuscript were written in Microsoft QuickBasic 7.1. The user will need a Quick Basic compiler to run the program. The files contained here do not serve as a general purpose simulator. New individual simulations can be created by appropriate modification of the files on the linked page. The program code is extensively referenced for clarity, but a few general comments may help orient users. The section “Trial Schedule” is where the user can turn stimuli on and off at different times. The program divides trials into discrete time slices represented by the counter “timeslices”. The number of time slices per trial can be set by the user. Experiments can be divided into up to four phases. Codes specific to simulations in the paper can be found here. Contact Steven Stout for any questions or suggestions concerning the simulation of the model. Steven C. Stout, Ph.D., B.C.B.A. Assistant Professor Department of Psychology Jacksonville State University 700 Pelham Road North Jacksonville, AL 36265-1602 Email link on Dr. Stout’s home page Dr. Miller is the corresponding author for questions about the manuscript: Ralph R. Miller, Ph.D. Distinguished Professor of Psychology Department of Psychology State University of New York at Binghamton Binghamton, NY 13902-6000, USA. |
|
Jacksonville State University Department of Psychology |
|
700 Pelham Road, Jacksonville, AL, 36265-1602. (256) 782-5402 |
|
A classic finding in conditioning research is that, if two conditioned stimuli have been paired with an unconditioned stimulus, they will each elicit less conditioned responding than a single conditioned stimulus that has been paired with the unconditioned stimulus. This phenomenon, called overshadowing, is representative of a host of phenomena usually referred to as cue competition in which cues appear to compete for control over behavior. However, there are also many observations of nontarget cues simultaneously present during target cue training facilitating the response to the target cue, phenomena that are called cue facilitation. To make the world utterly confusing, cue competition and cue facilitation can be generated with experimental protocols that look virtually identical. Most theories that predict competitive effects have little to say about mirror facilitative effects. The SOCR model attempts to address this discrepancy in an explicit manner. |