Retrieval from Semantic Memory
In some of my early work, I examined how information in semantic memory is organized (Balota & Coane, 2008) and what factors influence how and when we can access specific items (e.g., Coane & Balota, 2009, 2010, 2011; Coane et al., 2015). I have continued to use the tools of semantic priming – the facilitation observed when conceptually related information precedes a target for a response – to explore various forms of knowledge and how they influence, often outside of conscious awareness, task performance. For example, in collaboration with Dr. Mike Dacey (philosophy) at Bates College, we observed that primes depicting animals, whose faces appeared to convey happiness or sadness, facilitated response accuracy for word targets that were congruent with the animal’s apparent expression (Dacey & Coane, 2023, Frontiers). These findings suggest that anthropomorphic responses can occur rapidly and potentially outside of conscious awareness. Interestingly, although older adults tended to show larger priming effects, they reported less of a tendency to anthropomorphize on an explicit questionnaire, suggesting potential dissociations between implicit and explicit measures.
Relatedness Effects in Memory
The Deese-Roediger-McDermott (DRM) paradigm is a powerful tool for investigating memory errors and the underlying processes involved in what type of information individuals remember. The primary finding from this paradigm is that individuals reliably misremember the occurrence of a critical, non-presented item after studying lists of similar items. For example, after studying a list of words such as bed, rest, tired, awake, dream, etc., people often remember that sleep was also on the list, even when it was not. One account is that this occurs because activation spreads along pre-existing semantic and lexical connections, thereby increasing the familiarity or accessibility of the critical item. I am currently examining the limits of a spreading activation mechanism in accounting for memory errors and further clarifying what additional processes are critical for explaining this powerful memory illusion.
A new line of work that is currently underway involves examining the trajectory of phonological and semantic errors in second language learners.
Memory and Metamemory
This line of research involves examining how individuals rely on phenomenology to assess the contents of their memories and how they discriminate between different forms of retrieval failure (Coane & Umanath, 2019, JML; Umanath et al., 2023, CRPI) and retrieval success (Umanath & Coane, 2020, PPS; Coane et al., 2022, M&C). The new approach we have taken has been to rely on how individuals use natural language to describe their experiences with retrieval or retrieval failure. In one study (Umanath & Coane, 2020, PPS), we examined how both naïve and expert participants use the words “remember” and “know” in everyday language. Remembering was found to reflect the retrieval of specific events (episodic memory) and knowing reflected retrieval from the knowledge base, with broad consistency across participant groups.
In addition to assessing how participants use knowing and remembering to reflect their experiences of retrieval success, we (Coane & Umanath, 2019, JML) examined how participants express retrieval failures. Survey responses from naïve and expert participants indicated that not knowing reflects a failure in availability or storage (i.e., the information is not retrieved because it was never learned or stored), whereas not remembering reflects a retrieval failure or loss of information (i.e., the information is not retrieved because of a temporary failure to access it or because it was once known and has been forgotten).