Responding optimally with unknown sources of evidence (ROUSE) is a theory of short-term priming applied to associative, orthographic-phonemic, and repetition priming. In our studies, perceptual identification is measured with two-alternative forced-choice testing. ROUSE assumes features activated by primes are confused with those activated by the target. A near-optimal decision discounts evidence arising from such shared features. Too little discounting explains the finding that primed words were preferred after passive viewing of primes. Too much discounting explains the findings of reverse preference after active processing of primes. These preference changes highlight the need to use paradigms (like the present ones) capable of separating preferential and perceptual components of priming. Evidence of enhanced perception was found only with associative priming and was very small in magnitude compared with preference effects.

In most recognition models a decision is based on a global measure often termed familiarity. However, a response criterion is free to vary across lists varying in length and strength, making familiarity changes immeasurable. We presented a single list with a mixture of exemplars from many categories, so that the criterion would be unlikely to vary with length or strength of the category of the test item. False alarms rose with category length but not category strength, suggesting that familiarity does not change much with changes in strength of other items but grows when additional items are studied. The results were well fit by an extension of the search of associative memory (SAM) model presented by R. M. Shiffrin, R. Ratcliff, and S. E. Clark (1990).

Three experiments used the "list-before-the-last" free recall paradigm (Shiffrin, 1970) to investigate retrieval for context and the manner in which context changes. This paradigm manipulates target and intervening list lengths to measure the interference from each list, providing a measure of list isolation. Correct target list recall was only affected by the target list length when participants engaged in recall between the lists, whereas there were effects of both list lengths with other activities. This suggests that the act of recalling drives context change, thus isolating the target list from interference. Correspondingly, incorrect recall of intervening list items was affected only by the length of the intervening list when recall occurred between the lists, but was otherwise affected by both list lengths. Concurrent with these changes in context similarity, there were apparent changes in context retrieval, as indicated by the overall levels of target retrieval versus intervening recall. A multinomial model of sampling and recovery was implemented to assess the adequacy of this account and to quantify context similarity and context retrieval.

Three forced-choice perceptual word identification experiments tested the claim that transitions from positive to negative priming as a function of increasing prime duration are due to cognitive aftereffects. These aftereffects are similar in nature to perceptual aftereffects that produce a negative image due to overexposure and habituation to a stimulus. Each experiment tested critical predictions that come from including habituation in a dynamic neural network with multiple levels of processing. The success of this account in explaining the dynamics of repetition priming, associative-semantic priming, and forward masking effects suggests that habituation is a useful mechanism for reducing source confusion between successively presented stimuli. Implications are considered for visible persistence, repetition blindness, attention-based negative priming, attentional blink, inhibition of return, the negative compatibility effect, affect priming, and flanker preview effects.

No abstract

Perceptual input changes constantly in an unpredictable fashion, often changing before our somewhat sluggish perceptual systems have adequately processed this input. This can give rise to source confusion-how do we know whether a given perceptual activation is due to the current input, or a previous input that had yet to be completely processed? We propose that activity-dependent neural accommodation naturally limits this source confusion by suppressing items once they have been identified. We review behavioral paradigms from different literatures that measure the correlates of persistence and accommodation. Of the various accommodative mechanisms, we focus on synaptic depression, deriving a rate-coded expression that can be used to produce accommodating dynamics in any neural network with real valued activation. We implement this expression in a hierarchical model of perception termed, "a neural mechanism for responding optimally with unknown sources of evidence" (nROUSE). This model can be viewed as a more detailed version of the more abstract ROUSE model of Huber, Shiffrin, Lyle, and Ruys (2001), which produces accommodated levels of feature evidence through an optimal calculation. We apply nROUSE to three short-term priming experiments that manipulated prime duration.

Memory suppression is investigated with the no-think paradigm, which produces forgetting following repeated practice of not thinking about a memory [Anderson MC, Green C (2001) Nature 410:366 -369]. Because the forgotten item is not retrieved even when tested with an independent, semantically related cue, it has been assumed that this forgetting is due to an inhibition process. However, this conclusion is based on a single stage to recall, whereas global memory models, which produce forgetting through a process of interference, include both a sampling and a recovery stage to recall. By assuming that interference exists during recovery, these models can explain cue-independent forgetting. We tested several predictions of this interference explanation of cue-independent forgetting by modifying the think/nothink paradigm. We added a condition where participants quickly pressed enter rather than not thinking. We also manipulated initial memory strength and tested recognition memory. Most importantly, learning to quickly press enter produced as much cueindependent forgetting as no-think instructions. Demonstrating the adequacy of two-stage recall, a simple computational model (SAM-RI) simultaneously captured the original cue, independent cue, and recognition results.cued recall ͉ inhibition ͉ recall ͉ recognition ͉ computational model

The current study compared three models of recognition memory in their ability to generalize across yes/no and two-alternative forced-choice (2AFC) testing. The unequal-variance signal-detection model assumes a continuous memory strength process. The dual-process signal-detection model adds a threshold-like-recollection process to a continuous familiarity process. The mixture signal-detection model assumes a continuous memory strength process, but the old item distribution consists of a mixture of two distributions with different means. Prior efforts comparing the ability of the models to characterize data from both test formats did not consider the role of parameter reliability, which can be critical when comparing models that differ in flexibility. Parametric bootstrap simulations revealed that parameter regressions based on separate fits of each test type only served to identify the least flexible model. However, simultaneous fits of ROC data from both test types with goodness-of-fit adjusted using AIC successfully recovered the true model that generated the data. With AIC and simultaneous fits to real data, the unequal-variance signal-detection model was found to provide the best account across yes/no and 2AFC testing.