Wilding, E, Evans, L (2013 - 2014) Dissociating familiarity and conceptual priming with event-related potentials. Bial Foundation. £36,364.
This research proposal is designed to contribute to knowledge about the processes supporting human long-term memory judgments.
Identifying the appropriate information processing model for memory judgments is critical for: (i) understanding the contributions that distinct brain regions make to memory judgments, and (ii) predicting the kinds of memory problems that can ensue from specific patterns of brain damage. Event-related potential (ERP) data acquired during the test phases of retrieval tasks have been widely held to provide strong support for one particular memory model (Allan, Wilding & Rugg, 1998), but recent counter-claims challenge that view (e.g. Voss, Lucas & Paller, 2012).The purpose of the programme of work proposed here is to investigate the validity of the recent counter-claims.
The debate over the processes that support human long-term memory judgments, as well as the relationship between them, has spanned the cognitive psychology, cognitive electrophysiology and functional brain imaging literatures. One source of evidence that has been widely regarded as providing strong evidence for dual-process models of memory judgments is the findings in a series of event-related potential (ERP) studies (Wilding & Ranganath, 2011). According to dual-process models, the processes of recollection and familiarity make separable contributions to memory. Recollection is recovery of qualitative information about a prior encounter. Familiarity is a scalar strength signal that can support judgments of prior occurrence (Yonelinas, 2002).
The findings in studies of memory retrieval where ERPs have been acquired during the test phases have been held to support this dual-process model because two temporally and spatially distinct ERP modulations have been identified. Critically, these have varied according to experiment manipulations in a way that suggests one modulation - the mid-frontal ERP old/new effect - is an ERP index of familiarity, while the other – the left-parietal ERP old/new effect – is an index of recollection (Rugg et al., 1998).
A relatively recent challenge to this account claims that the mid-frontal old/new effect is not an index of familiarity (Voss, Lucas & Paller, 2010; 2012). Rather, it is an index of semantic priming. Priming is a change in performance as a function of a prior exposure, and is perhaps most commonly observed as a decrease in processing time when a stimulus is repeated. Semantic priming occurs when the performance change is observed for a stimulus because of the prior presentation of a semantically related stimulus.
The evidence that has been deployed in support of this alternative account of the functional significance of the mid-frontal ERP old/new effect is reviewed in section 3.2.1. For present purposes, the critical observation is that arguably the strongest means of assessing the functional significance of the mid-frontal 7/30 old/new effect is to obtain measures within the same experiment for semantic priming and for familiarity, observing with which one the mid-frontal old/new effect varies (Paller, Boehm & Voss, 2007). That is the approach taken in the two experiments described in this application, the intention being to adjudicate between the competing functional explanations that have been described above. This is an important issue: if the mid-frontal old/new effect is not an index of familiarity, then the ERP literature provides no support for dual-process accounts of human memory. Whilst the absence of this kind of support is far from fatal for dual-process accounts, it certainly weakens the evidence base considerably. Moreover, it would mean that ERPs are not an appropriate measurement instrument for assessing when, and to what degree, the processes of recollection and familiarity contribute to human memory judgments (e.g. Rosburg, Mecklinger & Frings, 2011; Azimian-Faridani & Wilding, 2004).