Poster F68, Tuesday, March 27, 8:00-10:00 am, Exhibit Hall C
Opposing mnemonic and decision-making biases in recognition memory judgments
Azara Lalla1, Anuya Patil1, Jennifer D Ryan1,2,3, Katherine Duncan1; 1Department of Psychology, University of Toronto, ON, Canada, 2Rotman Research Institute, Baycrest, Toronto, ON, Canada, 3Department of Psychiatry, University of Toronto, ON, Canada
Although memory assessments are generally assumed to faithfully quantify mnemonic processing per se, in practice they measure how people make decisions about memory. Yet, the ways in which decision-making biases shape performance is often overlooked and certainly understudied. Here, we uncovered one such bias by investigating how recent memory decisions impact recognition judgments. On a direct test of memory, we found that participants were more likely to judge an image as new after identifying an unrelated preceding image as old (p<0.05), and vice versa. The previous judgments did not affect subsequent accuracy; it only shifted the criterion that participants used to make their memory judgments. We replicated this bias across 4 conditions, but none could determine whether the sequential dependency in recognition judgments was in fact driven by a decision-making bias like the gambler’s fallacy or if they were due to a contrast effect in memory itself. In a final study, we removed the influence of decision-making by employing an eye-tracking paradigm to indirectly test memory for new and old faces. In stark contrast to our earlier findings, multiple eye-tracking metrics indicated that faces were treated as more novel when preceded by a novel face and as more familiar when preceded by a familiar face (p<0.05)—the opposite pattern to what was observed when participants made decisions about their memory. Taken together, these results suggest that decision-making biases have such a strong influence on recognition decisions that they can mask opposing biases in the mnemonic processes itself.
Topic Area: LONG-TERM MEMORY: Episodic