The Terrace simultaneous chaining paradigm

Terrace and his colleagues (Chen et al., 1997) have developed a simultaneous chaining paradigm (SCP) that breaks new ground in the study of animal learning. It also suggests a sharp contrast with human serial learning, and is pertinent to the on-going (ever-lasting) disagreement about item-to-item vs. item-to-position associations. I shall review this work and suggest the outline of a possible TODAM-like memory model using position rather than context vectors that might eventually provide a theoretical underpinning for this important work of Terrace. However, so far all I have been able to show is that this model can learn but it has not yet been developed to the point where we can estimate parameter values and fit the SCP data.