Home Director Members Research Publications Support

SAC (Source of Activation Confusion) Model of Memory

SAC is a unified model of implicit and explicit memory phenomena that spans both higher-level conceptual and lower-level perceptual effects. SAC's representation involves a rather generic semantic network of inter-associated concept nodes that vary in long-term strength. For example, there are nodes that represent letters (e.g., a, b, c), words (e.g. cat, climb, tree), context (e.g., experimental setting), sentences (e.g., The cat climbed the tree.), or parts of sentences, numbers (e.g., 5, 17, 31), operators (e.g., +, /, *), and whole problems (e.g., 17 + 31). It is assumed that concepts such as cat have as constituents the letters that spell the word, the phonemic features of how it is pronounced, the perceptual features of how it looks (orthography), as well as semantic information about the word. Although this model uses a localist, rather than a distributed representation, each concept is associated with a wide variety of features, a subset of which can activate the higher level node. A memory trace records the perceptual aspects of the experience as well as the conceptual aspects, and all aspects of a memory experience follow the same memory principles, regardless of whether the information is conceptual or perceptual. It is the detailed specification of how representations change with experience and how activation values are interpreted in particular situations that allows SAC to make specific, quantifiable predictions for many types of tasks. SAC has access to the activation values of its nodes which makes it easy for the model to make predictions concerning implicit memory experiments as well as predicting meta-cognitive judgments, such as Remember vs. Know judgments and feeling of knowing judgments. An important assumption of this model is that there is a single, unitary memory system that can explain the range of implicit and explicit memory phenomena. Although the hippocampus is thought to be responsible for binding in memory, that is not thought to distinguish implicit from explicit memory, per se. A recent focus in the lab is to investigate the role of encoding, especially the contributions of working memory, and the impact of familiarity on these processes. We test our ideas with a variety of methodologies, primarily behavioral but also some ERP (EEG), fMRI, and psychopharmacological (midazolam).

More Information About SAC

Papers and Topics using SAC:

Synthetic Amnesia and Memory

Midazolam is a benzodiazepine that is used routinely as a sedative during surgery. It causes temporary anterograde amnesia and when subjects are under the effects of the drug (at a modest dose of .03mg/kg of body weight), they typically forget much of the information that is presented to them. The effects of midazolam are short lived in comparison to other benzodiazepines, with amnestic effects having dissipated substantially at one hour after IV administration. We are currently using midazolam to examine implicit, explicit, and working memory, explore the mechanisms of interference, and understand the relationship between structure and function in the brain.

Papers on Synthetic Amnesia and Memory

Individual Differences in Strategy Adaptivity and Working Memory

Another project in the lab integrates my interests in skill learning and strategy choice. This work has led us to try to understand performance in an artificial air traffic controller task (funded by Office of Naval Research). The goals are to determine:

  • the strategies used initially and with more expert performance

  • how and when subjects shift their use of one strategy to another

  • whether there are individual differences in performance under cognitive load and in sensitivity to the shifting characteristics of the task

  • whether people vary in their sensitivity to shifting base rates in terms of the speed with which they alter their strategies

Papers on Working Memory

Implicit vs. Explicit (or unconscious vs. conscious) Strategy Adaptivity

In the 1980s, I discovered that subjects could quickly adapt the strategy that they were using to answer questions based on a shift in the base rate of the likelihood that a particular strategy would be useful, specifically a retrieval strategy or an inference/reasoning strategy. What I found more surprising is that although subjects were strongly affected by the shifting base rates, they were unaware of these base rates or the strategies that they were employing (they assumed that they always tried to search for the answer first). Since that time, we have found this same result, that subjects can quickly shift strategies to changing base rates in many other domains. We have also found that subjects are oblivious to the strategies they are using or these base rates.

From this pattern of results, I have concluded that much of our strategy selection (or what is sometimes called our metacognition) occurs implicitly or without awareness.

Papers associated with Implicit vs. Explicit (or unconscious vs. conscious) Strategy Adaptivity

Feeling of Knowing & Metacognition: Their Role in Strategy Selection for Question-Answering

Whether a person searches memory for the answer to a question, decides to deduce or induce the answer, or just responds "I don't know" without resorting to any other question answering strategy depends in part on whether the question and topic seems familiar. Some of my research has indicated that a very rapid feeling of knowing judgment influences an early evaluation of a question that gates the strategy selection to either attempt retrieval, try an alternative strategy, or give up. This view was considered implausible when first proposed. The dominant view at the time was that feeling of knowing the answer was determined by partial products from an early effort at retrieval of the answer. That is, the conventional wisdom, until recently, postulated that a person always tries to search for the answer before trying any other question-answering strategy.

Papers on Feeling of Knowing & Metacognition

Other Topics of Research and Their Associated Papers