John T. Wixted's research while affiliated with University of California and other places

... ; https://doi.org/10.1101/2023.01.04.522746 doi: bioRxiv preprint differ for words encoded during these two sessions as a function of word value. As the standard deviation of the raw ratings of studied words was significantly larger than that of lure words (the average ratio across 63 participants was 1.21 (SD = 0.207) and was significantly greater than 1, t(62) = 8.03, p < 0.001), and to take full advantage of our 6-point recognition confidence scale (which provides more information about memory strength than old/new judgments alone, we conducted a receiver operating characteristic (ROC) curve analysis using the unequal-variance signal-detection (UVSD) model (Brady et al., 2022;Mickes et al., 2007). This UVSD model yields memory sensitivity values (sometimes denoted as da) that are analogous to standard d' values that could be derived from the hit rates and false alarm rates alone, but are considered a superior measure because their derivation takes into account the entire distribution of confidence judgements to old and new items (Brady et al., 2022). ...

... Previous studies have often used K values to measure VWM capacity (Vogel et al., 2005;Ye et al., 2018), but other indexes (e.g., d ) have recently been suggested as potentially more valid measures of VWM in change detection tasks (Williams, Robinson, Schurgin, Wixted, & Brady, 2022). Therefore, for the VWM performance measurement, we analyzed both the K value and d (i.e., sensitivity), which can reflect the VWM performance of the participants. ...

... Speech (Supplementary Tables F, J) and music (Supplementary Tables G, K) are two classes of sound particularly important in human communication. A small proportion of hippocampal neurons respond selectively to specific spoken words (Urgolites et al., 2022), and in challenging listening conditions the distinctness of candidate word representations in left hippocampus is positively correlated with speech understanding (Blank et al., 2018;Fig. 4G). ...

... Moreover, researchers (Hung et al., 2017;Malbec et al., 2022) have argued that small sample sizes could lead to random outcomes and have even recommended a minimal sample size of 150 for correlational analyses (Hung et al., 2017). On the other hand, Wilson et al. (2022) suggested that theory-focused research should aspire to intermediate sample sizes, as too big sample sizes produce false positives, too. ...

... However, while performing the experiment for multiple trials, multiple traces of similar memory sets are formed in episodic memory. Due to effects of pro-and retroactive interference, this can lead to the result that traces of specific memory sets become inaccessible, even when cued with the presentation of the exact same set over and over again (Wixted, 2021(Wixted, , 2022. If the presentation of the repeated memory set cannot cue the retrieval of a previous episodic trace, no strengthening of that trace occurs; neither explicitly nor implicitly -rather, a new episodic-memory trace is created. ...

... Indeed, as Wixted and Wells highlight, in a dramatic and well-known eyewitness misidentification case in which Jennifer Thompson confidently but incorrectly identified Ronald Cotton as the man who raped her (Thompson-Cannino et al., 2010), Thompson's initial identification was indecisive and expressed with low confidence; it was only after receiving confirmatory feedback from police that Thompson expressed her identification with high confidence (see also Wells, 2020). To avoid the corrupting effect on memory of influences such as confirmatory feedback and misleading suggestions, Wixted et al. (2021) have recommended that a witness's memory should be tested only once. Given the possibly widespread impact of this recommendation on real-world cases, the progress made in this line of research is likely to be among the most consequential recent developments in research on memory sins in applied settings. ...

... Research shows that high eyewitness confidence provided immediately after a suspect identification from a fair lineup is indicative of suspect guilt (given the usual procedural safeguards; Wixted & Wells, 2017; for exceptions see Giacona et al., 2021), but memory for the perpetrator then becomes contaminated over time, resulting in high confidence expressed at trial not being indicative of suspect guilt (Wixted et al., 2022). Recent research also shows that potential jurors are not aware of this critical distinction between immediate and courtroom confidence, and will judge suspects as likely guilty when reading about high eyewitness confidence in a trial transcript, regardless of the timing (Key et al., 2022). ...

... The first challenge researchers face when linking performance across tasks, is measuring key hypothetical constructs. Although this is a well-known challenge for social science researchers in general (e.g., Borsboom et al., 2004;Brady et al., 2021;Kellen et al., 2020;Meehl, 1967;Regenwetter & Robinson, 2017;Rotello et al., 2015), it is also a challenge for researchers who study cognitive control in particular, because the dominant measurement approach called the subtraction method is known to suffer from major limitations. These limitations have been discussed in detail by other researchers and will be reviewed later in this article. ...

... Several recent papers have shown that filler similarity affects eyewitness ID performance (Carlson et al., 2019;Colloff, Wilson, Seale-Carlisle, & Wixted, 2021;Fitzgerald et al., 2013Fitzgerald et al., , 2015. Wells et al. (1993) proposed that lineups should contain a degree of propitious heterogeneity, such that fillers and suspect should not be too similar to each other in physical appearance. ...

... Although it is clear that fair lineups should be preferred, it is less clear exactly how many fillers are needed to maintain this advantage over showups. Only a few studies have examined showups in relation to multiple lineup sizes (e.g., Akan, Robinson, Mickes, Wixted, & Benjamin, 2021;Wooten et al., 2020) and so we will build upon this research by comparing showups, threeperson, and six-person lineups. We will not examine larger lineup sizes for two reasons. ...