Recognition memory and featural similarity between concepts: The pupil’s point of view

Abstract

Differences in pupil dilation are observed for studied compared to new items in recognition memory. According to cognitive load theory, this effect reflects the greater cognitive demands of retrieving contextual information from study phase. Pupil dilation can also occur when new items conceptually related to old ones are erroneously recognized as old, but the aspects of similarity that modulate false memory and related pupil responses remain unclear. We investigated this issue by manipulating the degree of featural similarity between new (unstudied) and old (studied) concepts in an old/new recognition task. We found that new concepts with high similarity were mistakenly identified as old and had greater pupil dilation than those with low similarity, suggesting that pupil dilation reflects the strength of evidence on which recognition judgments are based and, importantly, greater locus coeruleus and prefrontal activity determined by the higher degree of retrieval monitoring involved in recognizing these items.

Feature distinctiveness effects in language acquisition and lexical processing: Insights from megastudies

“…The facilitatory effect of Manhattan feature distinctiveness and inhibitory effect of Jaccard feature distinctiveness for concrete words suggest that, during the processing of concrete concepts, possessing feature information that distinguishes or discriminates between similar concepts (i.e., Manhattan distance) as well as feature information that is shared among many other concepts (i.e., Jaccard distance) are both important factors in semantic computation. This is consistent with much previous word meaning research, where both distinctive and shared features are important for identifying concepts (Cree et al, 2006;Montefinese, Vinson, & Ambrosini, 2018;Montefinese, Zannino, & Ambrosini, 2015;Taylor, Devereux, Acres, Randall, & Tyler, 2012). On the other hand, possessing feature information that does not distinguish or discriminate between similar concepts (i.e., Manhattan distance) as well as feature information that is not shared among many other concepts (i.e., Jaccard distance) appear to be important in the semantic processing of abstract concepts.…”

Pupil size changes signal hippocampus-related memory functions

“…Finally, 'old' responses to foils, which are semantically related to target items also show sensitivity to objective and subjective familiarity: 'old' responses to such lure items fell in between correct 'old' and correct 'new' responses to targets and foils, respectively 54 . Similar result was reported for lures with low degree of similarity to targets, whereas 'old' responses to lures with high similarity to targets were indistinguishable from 'old' responses to targets 58 . This pattern of results suggests that both subjective familiarity (the act of responding with 'old') and objective familiarity (overlap between target and lure representations) contributes to PD magnitude.…”

You can believe your eyes: measuring implicit recognition in a lineup with pupillometry

“…Pupils are larger when memory strength is greater (Brocher & Graf, 2016;Otero et al, 2011;Papesh et al, 2012). They may also reflect the experience of recognition (Otero et al, 2011), and the strength of evidence on which recognition is based (Montefinese et al, 2018).…”

“…Pupillometry has also proved useful in indexing memory strength, as pupils have been shown to be larger when retrieving items associated with greater memory strength (Brocher & Graf, 2016;Otero, Weekes, & Hutton, 2011;Papesh, Goldinger, & Hout, 2012), and they also appear to reflect the experience of recognition (Otero et al, 2011). Therefore, pupil size may also reflect the strength of recognition evidence (Montefinese, Vinson, & Ambrosini, 2018). Pupillometry also appears to be useful for measuring implicit memory, as pupillary changes occur in the absence of an overt response (van Rijn, Dalenberg, Borst, & Sprenger, 2012), and can even occur despite efforts to deceive (Heaver & Hutton, 2011).…”