Search

During his distinguished career, Bill Hockley contributed to memory research in many ways, with work characterized by rigorous and innovative experimental designs. One of the areas he has explored is that of memory for associative information. We echo this interest here and attempt to emulate his careful experimental attitude. We report four experiments which examined how previously established links can support the development of new episodic associations. More specifically, we tested the idea that sound-symbolism links can support learning of new associations. Sound-symbolism links are relationships between phonemes and object characteristics that participants find natural-even if they have never encountered the items before. For instance, the nonword "honulo" is more readily seen to refer to a shape with curved contours than to a shape that has sharp angles. In Experiment 1, 70 participants studied three pairs and their memory for the associations between the members of each pair was tested in a paired-recognition task. Results demonstrate that sound-symbolism associations support the learning of new associations. Experiment 2 confirmed that the effect is replicated in a between-participants design. In Experiment 3, we replicated the findings with a 30-s filled interval between presentation and test, and in Experiment 4, we extended the delay to 2 min, establishing that the pattern is also found with a paradigm more typical of episodic memory. The results are discussed in terms of the importance of associative memory, while referring to some of the ideas Bill Hockley championed in his own work. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

In backward immediate serial recall, participants recall lists of items immediately after their presentation by beginning with the last presented item and ending with the first presented one. Despite the similarities with forward recall in which participants recall the items from the first to the last presented, benchmark memory phenomena reliably found in forward recall are not constantly observed in backward recall. Here, we proposed a new framework called the encoding-retrieval matching (ERM) hypothesis to account for backward recall. The ERM retains the main features of the visuospatial hypothesis and the item-order trade-off hypothesis, the two dominant accounts of backward recall. According to the ERM, output modality and foreknowledge of recall direction influence the availability of visuospatial representations and the weight devoted to item and order processing. We tested the ERM with irrelevant speech, a well-known working memory factor disrupting forward recall. In two experiments, we manipulated recall direction (forward vs. backward), irrelevant speech (control vs. irrelevant speech), and response modality (manual vs. oral). As predicted by the ERM, when recall direction was unpredictable in Experiment 1, the magnitude of the irrelevant speech effect was larger in backward manual recall than in backward oral recall. In Experiment 2, recall direction was predictable. As predicted by the ERM, in backward recall, the irrelevant speech effect was reduced with a manual response and absent with an oral response. We concluded that ERM effectively accounts for the complex interplay between response modality, foreknowledge of recall direction, and benchmark memory effects in backward recall. (PsycInfo Database Record (c) 2021 APA, all rights reserved).

The

Over the last decades, the extralinguistic benefits of bilingualism have been intensively debated. The current study was aimed at clarifying whether bilingualism speeds attentional disengagement. Reflecting faster disengagement, Mishra, Hilchey, Singh, and Klein (2012) observed an earlier onset of inhibition of return (IOR) for high than for low-proficient bilinguals. In contrast, Hernandez, Costa, Fuentes, Vivas, and Sebastian-Galles (2010) failed to find any difference between bilinguals and monolinguals. We investigated the source of this discrepancy, while improving methodology by using a large sample composed of 100 Canadians, objective assessments of second language skills (Nelson-Denny Reading test), and controlling for nonverbal intelligence, age, sex, and video-gaming. Results were analyzed with self-report and objective measures of second language proficiency as well as dichotomous and continuous measures. Compared to less proficient bilinguals, highly proficient bilinguals tended to respond faster overall, hinting at an executive processing advantage. However, contrary to Mishra et al.'s findings, bilingual proficiency did not affect either the onset of IOR or magnitude of IOR. (PsycINFO Database Record

In immediate serial recall, it is well established that performance is influenced by lexical factors such as imageability and word frequency. However, when participants are asked to recall the to-be-remembered items in their reverse order, known as backward recall, lexical factors produced contradictory findings. In 4 experiments, we tested the role of response modality in modulating the effects of imageability and word frequency in backward recall. The magnitude of the 2 tested lexical factors was only slightly reduced in backward compared with forward recall when participants responded manually. Critically, when participants recalled words orally, the magnitude of the lexical factors was severely reduced in backward recall. It is concluded that response modality has a strong influence on the memory processes involved in backward recall. (PsycINFO Database Record

The word-length effect, the finding that lists of short words are better recalled than lists of long words, is 1 of the 4 benchmark phenomena that guided development of the phonological loop component of working memory. However, previous work has noted a confound in word-length studies: The short words used had more orthographic neighbors (valid words that can be made by changing a single letter in the target word) than long words. The confound is that words with more neighbors are better recalled than otherwise comparable words with fewer neighbors. Two experiments are reported that address criticisms of the neighborhood-size account of the word-length effect by (1) testing 2 new stimulus sets, (2) using open rather than closed pools of words, and (3) using stimuli from a language other than English. In both experiments, words from large neighborhoods were better recalled than words from small neighborhoods. The results add to the growing number of studies demonstrating the substantial contribution of long-term memory to what have traditionally been identified as working memory tasks. The data are more easily explained by models incorporating the concept of redintegration rather than by frameworks such as the phonological loop that posit decay offset by rehearsal. (PsycINFO Database Record

The serial position function, with its characteristic primacy and recency effects, is one of the most ubiquitous findings in episodic memory tasks. In contrast, there are only two demonstrations of such functions in tasks thought to tap semantic memory. Here, we provide a third demonstration, showing that free recall of the prime ministers of Canada also results in a serial position function. Scale Independent Memory, Perception, and Learning (SIMPLE), a local distinctiveness model of memory that was designed to account for serial position effects in episodic memory, fit the data. According to SIMPLE, serial position functions observed in episodic and semantic memory all reflect the relative distinctiveness principle: items will be well remembered to the extent that they are more distinct than competing items at the time of retrieval. Keywords: serial position function, semantic memory, episodic memory, recency effect (ProQuest: ... denotes formulae omitted.) One ubiquitous finding in episodic memory tasks is the serial position function. Recent items are well remembered (the recency effect), the first few items are also well remembered (the primacy effect), but midlist items are less well recalled. Such serial position functions are readily observable over timescales ranging from milliseconds (Neath & Brown, 2006) to minutes (Bjork & Whitten, 1 974) to days and weeks (Glenberg, Bradley, Kraus, & Renzaglia, 1983) to months (Huttenlocher, Hedges, & Prohaska, 1992) and even to years (Sehulster, 1989). While such functions are less commonly observed in memory tasks thought to tap semantic memory, there are two such reports, one involving recall of US presidents (Roediger & Crowder, 1976) and the other verses of well-known hymns (Maylor, 2002). ' In this paper, we report a third example of a serial position function from semantic memory, free recall of Canadian prime ministers. We then test whether a general explanation of serial position functions that applies to both episodic and semantic memory systems fits these new data. To the extent that the model fits, it offers further evidence for the view that there exist common memory principles that apply regardless of the hypothetical underlying memory system (Surprenant & Neath, 2009). Serial Position Functions in Semantic Memory Roediger and Crowder (1976) reported a standard-looking serial position function when students were asked to recall the presidents of the United States.2 There was a substantial primacy effect, a pronounced recency effect, and quite poor recall of middle presidents with the exception of Lincoln. This latter finding was likened to a von Restorff or isolation effect. Similar results were reported by Crowder (1993), who updated the results to include Carter and Reagan. Of interest, those presidents in the recency portion of the list in 1976 (e.g., Nixon and Ford) were less well recalled in 1993 than those presidents who replaced them in that part of the list (e.g., Reagan and Bush). Healy and colleagues (Healy, Havas, & Parker, 2000; Healy & Parker, 2001) have replicated this result when including Clinton. The second demonstration of a serial position function in semantic memory was reported by Maylor (2002). She presented churchgoers with six verses of well-known hymns in random order and asked them to rearrange the verses to reflect the correct order. Importantly, she had a control group of subjects who were unfamiliar with the hymns. Whether one looked at just the proportion correct of the churchgoers or the proportion correct adjusted for level of performance of the nonchurchgoers, there was a primacy effect, a recency effect, and the midlist items were poorly recalled. SIMPLE SIMPLE (Scale Independent Memory, Perception, and Learning) is a relative distinctiveness model of memory that was originally designed to explain serial position effects in episodic memory tasks (Brown, Neath, & Chater, 2007) but it has also been fit to both the presidential data (Neath, 2010) and the hymn data (Neath & Brown, 2006). …

When participants read a text while searching for a target letter, they are more likely to miss the target letter embedded in frequent function words than in less frequent content words. This effect is usually observed with a text displayed normally, for which it has been found that frequent function words are fixated for a smaller amount of time than less frequent content words. However, similar pattern of omissions have been observed with a rapid serial visual presentation procedure in which words appear one at a time. These parallel results would demonstrate that fixation duration per se is not the proximal cause of the missing-letter effect only if eye movements are not made during the rapid serial visual presentation procedure. Therefore, the authors performed eye monitoring during the rapid serial visual presentation procedure. Results revealed that, with a rapid serial visual presentation procedure, participants fixated function and content words for almost the entire presentation duration. It is concluded that eye movements are not the proximal cause of the missing-letter effect.

Immediate Serial Recall, Word Frequency, Item Identity and Item Position MARIE POIRIER and JEAN SAINT - AUBIN, Universite Laval Abstract Eighteen subjects completed an immediate serial recall task, where the to - be - recalled lists consisted of either high, medium, or low - frequency items. Moreover, lists were either phonologically similar or distinct. Results showed that increasing frequency enhanced item information recall but had no effect on order recall. Conversely, increasing phonological similarity had a detrimental effect on order recall but no significant effect on item recall. It is argued that both effects reflect retrieval processes where degraded representations are reconstructed on the basis of long - term knowledge: Low - frequency words have reduced accessibility, lowering the probability of correct reconstruction, and phonologically similar items are more easily confused with other recall candidates. In immediate serial recall (ISR), subjects must recall a short list of items immediately after their presentation. It is now well established that lists of high - frequency words produce better performance than lists of low - frequency words (Engle, Nations, & Cantor, 1990; Gregg, Freedman, & Smith, 1989; Kausler & Pucket, 1979; Roodenrys, Hulme, Alban, Ellis, & Brown, 1994; Watkins, 1977; Watkins & Watkins, 1977). The aim of the work reported here was to answer a further question: Does frequency enhance memory for an item's occurrence, memory for an item's position within a list, or both (Whiteman, Nairne, & Serra, 1994)? Recent theoretical proposals can be seen as arguing either way. For example, Roodenrys et al. (1994) proposed a two - stage view of frequency effects, where partially degraded traces are retrieved from a phonological short - term store and are then subjected to a "deblurring" or pattern completion process. They suggest that this process depends on long - term representations, and that high - frequency words are better recalled because of their greater accessibility. In other words, frequency affects the probability of a successful deblurring attempt. A similar account of lexicality effects in ISR has been proposed (Hulme, Maughan, & Brown, 1991; Schweickert, 1993), and it has also been usefully applied to the semantic category advantage (Poirier & Saint - Aubin, 1995). This type of proposal predicts an effect of frequency on item identity (Poirier & Saint - Aubin, 1995; see also Estes, 1991). All else being equal, the probability of successful deblurring will be greater for high - frequency items, leading to better item identity recall. The main objective of the experiment reported here was to test this prediction. In contrast to accounts emphasizing item information, a number of proposals attribute the frequency advantage to better order information encoding, the hypothesis being that high - frequency lists entail stronger inter - item links (Deese, 1960; Sumby, 1963; Whiteman et al., 1994). Notably, Lewandowsky and Murdock (1989) successfully applied the Theory of Distributed Associative Memory to frequency effects in ISR by assuming that higher frequency lists produced stronger inter - item associations. A relevant empirical test of this class of proposals was recently conducted by Whiteman et al. (1994). They examined the effect of word frequency on a reconstruction task. After list presentation, subjects were provided with the items in a new random order and asked to reconstruct the original presentation sequence. It follows that the task is mainly one of order information recall (Whiteman et al., 1994). Based on the hypothesis that high - frequency enhances order memory, superior performance is expected in the high - frequency condition. The Whiteman et al. (1994) results did not support this prediction: Word frequency did not influence performance. However, the reconstruction task used by Whiteman et al. …