Can a chaining model account for serial recall?

Henson (1996) has argued that several results including fillin effects, patterns of protrusions and performance on lists of alternating similar and dissimilar items (the sandwich effect) preclude a model of serial recall that relies on chaining associations between items. However, this conclusion is at odds with other data showing that serial recall improves dramatically when study lists approximate language at the letter and word levels and also is improved when circular lists that maintain chaining information, but confound positional information are repeated. In this paper, I demonstrate that the objections to chaining models can be overcome if one assumes that associations act as constraints on a whole of list resolution process, rather than acting in a purely feedforward fashion.