Abstract
Recent work has revealed that participants can frequently fail to remember information (e.g., the identity of a letter) even immediately after processing it (finding the letter in a set of numbers to report its color). This ‘attribute amnesia’ effect (Chen & Wyble, 2015) poses a problem for our intuitions about conscious awareness; how can we not know the identity of a target we just consciously processed and correctly identified as a letter? Chen and Wyble argue that the target identity is perceptually encoded, but not consolidated into memory unless subjects are required to hold it in mind for a few hundred milliseconds. An alternative to this distinction between perceptual encoding and consolidation is to consider whether participants ever consciously process the identity of the letter at all. That is, the connection between a visual stimulus and the concepts activated in its apprehension may not be straightforward; depending on context and task demands, different levels of relevant concepts may be activated. Thus, when searching for a letter amongst a set of non-letter distractors, it may be that only the well-learned concept ‘letter’ is ever consciously activated, rather than the specific letter identity. To test this, we compared well-established target categories, such as ‘letter’ and ‘number’, with less well-established categories, such as ‘letter from the second half of the alphabet’, using the paradigms of Chen and Wyble (2015). We found that the attribute amnesia effect for target identity disappeared when the target category was not well-established – that is, reporting the color of the ‘letter from the second half of the alphabet’ reliably led to memory for letter identity, though reporting the color of a ‘letter’ did not. This result argues for a closer look at the role of conceptual activation and conceptual short-term memory (Potter, 1976) in perception and working memory.
Acknowledgement: NSF BCS-1829434 to TFB