Furthermore, their intact performance on the Low Interference conditions, particularly the second Low Interference condition, demonstrates that their deficits were specific to the buildup of interfering features, rather than fatigue or generic task-practice BTK inhibitor effects. The hippocampal cases were not impaired on any condition (all t(7) < 0.4, p > 0.3). To address the potential concern that differences in task difficulty across conditions could have confounded our results, we analyzed the accuracy and reaction time data of control participants (shown in Figures 5 and 6; Tables S1 and S7; all reported t tests are two-tailed). Importantly, the planned interaction contrast revealed

no greater difference in d′ between High Ambiguity and Low Ambiguity Objects than between Difficult and Easy Size (the interaction was not significant in experiment 2, t(19) = 1.1, p = 0.3, and was driven by a bigger drop in performance for Difficult than Easy size conditions in experiments 1 and 3, both t > 2.0, p < 0.06). In experiment 4, the condition on which the MTL patients were impaired (High Interference) was not the condition that controls found to be the most difficult: the High Interference condition was matched in difficulty to Low Interference 2 (t(21) = 0.3, p = 0.8) and significantly easier than the selleck chemical Low Interference 1 (t(21) = 3.1, p < 0.01). These results suggest that our observed

eye movement patterns (expt 1), fMRI effects of feature ambiguity (expt 2), and patient deficits (expts 3–4) were not due to global differences in task difficulty. In terms almost of reaction times, the increase in RTs for High Ambiguity versus Low Ambiguity Objects was significantly greater than the increase for Difficult versus Easy Size in experiments 1–3 (a trend in expt 1: t(15) = 1.9, p = 0.07; expts 2 and 3: both t > 2.2, p < 0.05). In experiment 4, reaction times for the High Interference condition were significantly longer relative to the second Low Interference condition (t(21) = 3.0, p < 0.01),

but were not significantly different from the first Low Interference condition (t(21) = 1.5, p = 0.2). These results suggest that at least for experiment 4, differences in reaction times cannot explain the patients’ deficits. Nonetheless, the finding that in experiments 1–3 the High Ambiguity Object conditions were associated with longer reaction times relative to the Size Control conditions merits further consideration in light of the idea that working memory demands may have differed across conditions. Several studies have reported impairments of short-term memory in amnesia (e.g., Hannula et al., 2006, Nichols et al., 2006, Olson et al., 2006, Warren et al., 2010, Warren et al., 2011 and Warren et al., 2012), and neuroimaging studies have observed hippocampal activity in tasks typically considered to assess short-term memory (e.g., Cabeza et al., 2002, Cashdollar et al., 2009, Hannula and Ranganath, 2008, Karlsgodt et al.