Journal of memory and language最新文献

Accurate word recognition is facilitated by context. Some relevant context, however, occurs after the word. Rational use of such “right context” would require listeners to have maintained uncertainty or subcategorical information about the word, thus allowing for consideration of possible alternatives when they encounter relevant right context. A classic study continues to be widely cited as evidence that subcategorical information maintenance is limited to highly ambiguous percepts and short time spans (Connine et al., 1991). More recent studies, however, using other phonological contrasts, and sometimes other paradigms, have returned mixed results. We identify procedural and analytical issues that provide an explanation for existing results. We address these issues in two reanalyses of previously published results and two new experiments. In all four cases, we find consistent evidence against both limitations reported in Connine et al.’s seminal work, at least within the classic paradigms. Key to our approach is the introduction of an ideal observer framework to derive normative predictions for human word recognition expected if listeners maintain and integrate subcategorical information about preceding speech input rationally with subsequent context. We test these predictions in Bayesian mixed-effect analyses, including at the level of individual participants. While we find that the ideal observer fits participants’ behavior better than models based on previously proposed limitations, we also find one previously unrecognized aspect of listeners’ behavior that is unexpected under any existing model, including the ideal observer.

Real-world environments are complex, demanding a diverse set of cognitive functions such as attention and working memory (WM) to perform adaptive behaviors. However, exogenous attention, characterized as automatic and involuntary, has primarily been studied by focusing on spatial perception. In particular, the ability of pure exogenous retro-cues to select and prioritize not only spatial locations, but also novel stimulus–response (S-R) bindings held in WM remains largely unexplored. Here, in two experimental series, we provide evidence that pure exogenous non-predictive retro-cues can select not only space, but also associated S-R bindings held in WM. Additional evidence from a drift–diffusion model hinted at the possibility that the mechanisms through which exogenous attention selects and prioritizes WM contents depend, at least partially, on the hierarchical relevance of the different dimensions encoded within a specific representation. These results highlight the relationship between pure exogenous attention and complex WM contents and shed light on current theoretical debates about the interaction of attention, memory, and action.

The long-term negative effect of semantic retrieval on the subsequent accessibility of related material has been extensively studied in separate memory and language production literatures. Though ostensibly studying the same phenomenon, these literatures have remained separated by different framings and methodologies. We argue for integration of the two research streams in an adaptive learning perspective and present a bridging experiment as a proof of concept of this approach. The experiment implemented a multiphase retrieval-induced forgetting (RIF) design (with generation and memory assessment phases) in combination with the use of naming latency measures and the temporal analysis of interference featured in language production research. The generation phase, typically unanalyzed in the memory literature, examined generation time for category-stem completions as a function of ordinal positions of related items. There was strong cumulative interference in generation latencies in the first pass through the structured list, showing that memory is already affected in this phase. After a retention interval, accessibility of new items from previously activated categories, and unactivated controls, was assessed using continuous picture naming rather than aggregate memory measures. Crucially, there was a picture naming cost to previously activated (but not generated) category members relative to the control condition, a RIF effect. This cost was supervenient on new cumulative interference and was evident only in the beginning of the assessment phase, underlining the value of the positional analyses. The findings add important detailing to the processes underlying retrieval-induced costs in memory research while also showing that retrieval-induced semantic interference transfers from stem-completion to picture naming retrieval tasks. This format-independence is consistent with a conceptual basis of semantic interference but does not preclude a locus of adaptive learning in conceptual-lexical links. Overall, we show that the memory and language production fields indeed provide different but complementary perspectives on the same semantic interference phenomenon. Combining the fields promises to be productive.