Journal of memory and language最新文献

It has been proposed that the order in which words are prepared for production depends on the speaker’s language. When producing the translation equivalent of the small cat, speakers of German or Dutch select the gender-marked determiner at a relatively early stage of production. Speakers of French or Italian postpone the encoding of a determiner or adjective until the phonological form of the noun is available. Hence, even though the words are produced in the same order (e.g., die kleine Katze in German, le petit chat in French), they are not planned in the same order and might require different amounts of advanced planning prior to production onset. This distinction between early and late selection languages was proposed to account for the observation that speakers of Germanic and Slavic languages, but not of Romance languages, are slower to name pictures in the context of a distractor word of a different gender. Meta-analyses are conducted to provide the first direct test of this cross-linguistic difference and to test a prediction of the late selection hypothesis. They confirm the existence of the gender congruency effect in German/Slavic languages and its absence in Romance languages when target and distractor words are presented simultaneously. They do not allow confirming the hypothesis that in the latter languages, a similar effect emerges when the presentation of the distractor is delayed. Overall, these analyses confirm the cross-linguistic difference but show that the evidence available to date is not sufficient to confirm or reject the late selection hypothesis as an explanation of this difference. We highlight specific directions for future research.

There is a long-standing controversy around the question of whether referent predictability affects pronominalization: while there are good theoretical reasons for this prediction (e.g., Arnold, 2008), the experimental evidence has been rather mixed.

We here report on three highly powered studies that manipulate a range of factors that have differed between previous studies, in order to determine more exactly under which conditions a predictability effect on pronominalization can be found.

We use a constrained as well as a free reference task, and manipulate verb type, antecedent ambiguity, length of NP and whether the stimuli are presented within a story context or not. Our results find the story context to be the single important factor that allows to elicit an effect of predictability on pronoun choice, in line with (Rosa and Arnold, 2017; Weatherford and Arnold, 2021). We also propose a parametrization for a rational speech act model, that reconciles the findings between many of the experiments in the literature.

Theories of speech production have proposed that in contexts where multiple languages are produced, bilinguals inhibit the dominant language with the goal of making both languages equally accessible. This process often overshoots this goal, leading to a surprising pattern: better performance in the nondominant vs dominant language, or reversed language dominance effects. However, the reliability of this effect in single word production studies with cued language switches has been challenged by a recent meta-analysis. Correcting for errors in this analysis, we find that dominance effects are reliably reduced and reversed during language mixing. Reversed dominance has also consistently been reported in the production of connected speech elicited by reading aloud of mixed language paragraphs. When switching, bilinguals produced translation-equivalent intrusion errors (e.g., saying pero instead of but) more often when intending to produce words in the dominant language. We show this dominant language vulnerability is not exclusive to switching out of the nondominant language and extends to non-switch words, linking connected speech results to patterns first reported in single-picture naming studies. Reversed language dominance is a robust phenomenon that reflects the tip of the iceberg of inhibitory control of the dominant language in bilingual language production.

Prior testing potentiates new learning, an effect known as test-potentiated new learning (TPNL). Research using lists of related words has established that testing, by free recall, also increases semantic clustering of later recall output. It has been suggested that this is evidence that testing induces a strategy change in encoding and retrieval towards greater conceptual organisation. The current research evaluated whether this conceptual strategy change explains TPNL in three experiments. We found a) that a retrieval task that did not increase semantic clustering (list discrimination) consistently produced TPNL, and b) that factors (word-relatedness and list structure) that influenced the amount of semantic clustering had no effect on the magnitude of TPNL. These results suggest that conceptual strategy change is neither necessary nor sufficient for TPNL and is more likely to be an effect of testing, rather than a cause of TPNL.

In sentence comprehension, what are the cognitive constraints that determine number agreement computation? Two broad classes of theoretical proposals are: (i) Representation distortion accounts, which assume that the number feature on the subject noun gets overwritten probabilistically by the number feature on a non-subject noun, leading to a non-veridical memory trace of the subject noun; and (ii) The cue-based retrieval account, a general account of dependency completion processes which assumes that the features on the subject noun remain intact, and that processing difficulty is only a function of the memory constraints on dependency completion. However, both these classes of model fail to account for the full spectrum of number agreement patterns observed in published studies. Using 17 benchmark datasets on number agreement from four languages, we implement seven computational models: three variants of representation distortion, two cue-based retrieval models, and two hybrid models that assume both representation-distortion and retrieval. Quantitative model comparison shows that the best fit is achieved by a hybrid model that assumes both feature distortion (specifically, feature percolation) and cue-based retrieval; numerically, the second-best quantitative fit was achieved by a distortion-based model of number attraction that assumes grammaticality bias during reading. More broadly, the work furnishes comprehensive evidence to support the idea that cue-based retrieval theory, which aims to be a general account of dependency completion, needs to incorporate a feature distortion process.

Speakers frequently reuse earlier encountered structures. A long-standing view in language production research is that this structural priming is driven by the persistence of abstract syntax, independent from unordered, conceptual representations. However, evidence has been building that non-syntactic information can also influence structural choice. Here we examined whether and how the syntactic priming of relative clause structures might interact with the priming of the conceptual category order of adjectives in noun phrase production. Study 1 found that speakers are more likely to produce relative clause structures (spotted bow that’s green) after having heard relative clause structures (striped lock that’s blue) as opposed to an alternative structure (striped blue lock), and they also tended to repeat the conceptual order of the prime, with more pattern-first orders after pattern-first primes than after color-first primes. Critically, we found larger syntactic priming when the conceptual order of the prime persisted more in the target and larger conceptual order priming when the syntactic structure of the prime persisted more in the target. Studies 2 and 3 found that conceptual category order priming can be enhanced by adjective overlap as well as noun overlap between prime and target, whereas syntactic priming can only be enhanced by noun overlap. These results supported the interactive priming account: Although the syntactic structure and the conceptual order are represented at different levels and hence can be activated independently, the link between them is also primed, which enhances priming at both levels.

Imageability – the ease of generating a mental image for a word – has been commonly used as a predictor of word recognition but its effects are highly variable across the literature, raising questions about the robustness and stability of the construct. We compared six existing imageability norms in their ability to predict RT and accuracy in lexical decision and word naming across thousands of words. Results showed that, when lexical and sensorimotor sources of variance were partialled out, imageability predicted little unique variance in word recognition performance and effect sizes varied greatly between norms. Further analysis suggested that such heterogenous effect sizes are likely due to inconsistent strategies in how participants interpret and rate imageability in norming studies, despite consistent instructions. Our findings suggest that the ease of generating a mental image for a word does not reliably facilitate word recognition and that imageability ratings should be used with caution in such research.

Spelling-to-sound translation in English is particularly complex for vowels. For example, the pronunciations of ‹a› include the long vowel of ‹paper› and ‹sacred› and the short vowel of ‹cactus› and ‹happy›. We examined the factors that are associated with use of long versus short vowels by conducting analyses of English disyllabic words with single medial consonants and consonant sequences and three behavioral studies in which a total of 119 university students pronounced nonwords with these structures. The vocabulary analyses show that both the number of medial consonants and their syllabification influence vowel length. Participants were influenced by these aspects of context, some of which are not explicitly taught as a part of reading instruction. Although these results point to implicit statistical learning, participants produced fewer long vowels before single medial consonants than anticipated based on our vocabulary statistics for spelling-to-sound correspondences in disyllabic words. Participants also produced more long vowels before two identical consonant letters than anticipated given these statistics. We consider the reasons for these outcomes, and we also use the behavioral data to test two models of spelling-to-sound translation.

In this study, we replicated and extended Experiment 1 of Palmeri et al. (1993) in two experiments. Using the continuous recognition memory paradigm, we investigated effects of a demographically heterogeneous set of talkers varying across race, gender, and regional accent (Exp. 1) and effects of two demographically homogeneous sets of talkers (8 identifiably white male or 8 identifiably Black male talkers) across two listener populations (white and Black listeners) (Exp. 2). Words repeated in the same voice were recognized more quickly and accurately than words repeated in a different voice in both experiments, as found in the original study. This pattern is extremely robust. However, we also found differences across talker conditions, number of voices, lag, false alarms, and d’ that differ from the original study (Exp. 1). In addition, we found effects of talker, talker context, and listener population suggesting that social ideologies and experiences greatly influence the encoding of and memory for spoken words (Exp. 2).