The influence of word frequency on recency effects in directed free recall

Leading theoretical explanations of recency effects are designed to explain the reported absence of a word frequency effect on recall of words from recency serial positions. The present study used a directed free-recall procedure (J. J. Dalezman, 1976) and manipulated the frequency composition of the word lists (pure and mixed). Overall, with pure lists, a greater proportion of high-frequency (HF) words were recalled than low-frequency (LF) words, and with mixed lists, a greater proportion of LF words were recalled than HF words. Of importance, this recall advantage for one frequency over the other as a function of list composition was evident across the last three serial positions, indicating an influence of word frequency on recency effects that is dependent on the frequency composition of the lists. These results challenge one of the major assumptions on which several theories of recency effects have been based.     

词频、易接近性和词表序列成分对近因效应影响的实验研究

本研究目的探讨词频、易接近性和词表序列成分对近因效应的影响及其原因。两个实验以不同词频和不同易接近性的中文词为材料,运用直接自由回忆程序和操纵词表的组成成分,探讨近因效应的特点。实验结果表明：词频对近因效应的作用受词表序列成分的影响,对于纯的词表,高频词回忆的比例比低频词更高,对于混合词表,低频词回忆的比例比高频词更高,但是词的易接近性不影响近因效应且不受词表序列成分影响。文章最后对近因效应的性质进行了讨论。     

工作记忆中汉语反词长效应机制

工作记忆中的反词长效应(reverse word-length effect)指在对长词和短词混合的词表进行即时序列回忆时, 独立长词回忆成绩优于独立短词的现象。以汉字词语为材料通过3个实验探讨反词长效应的机制。实验1采用纯粹词表和长短词混合词表, 既得到纯粹词词长效应, 也得到独立词反词长效应。实验2削弱了长短词之间的词长差异, 结果独立词反词长效应消失, 且独立词回忆成绩优于纯粹词。实验3设计了视觉延迟条件, 得到与实验1类似的结果, 只是独立词反词长效应有所削弱。三个实验的结果并不一致, 无法用现有的语音回路理论或SIMPLE理论进行很好地解释, 理论的整合与创新显得非常重要。因此, 提出多重编码以既相互竞争又相互补充方式进行平行加工的观点进行更完整地解释。     

Part-list cuing and the dynamics of false recall

False recall of an unpresented critical word after studying its semantic associates can be reduced substantially if the strongest and earliest-studied associates are presented as part-list cues during testing (Kimball & Bjork, 2002). To disentangle episodic and semantic contributions to this decline in false recall, we factorially manipulated the cues’ serial position and their strength of association to the critical word. Presenting cues comprising words that had been studied early in a list produced a greater reduction in false recall than did presenting words studied late in the list, independent of the cues’ associative strength, but only when recall of the cues themselves was prohibited. When recall of the cues was permitted, neither early-studied nor late-studied cues decreased false recall reliably, relative to uncued lists. The findings suggest that critical words and early-studied words share a similar fate during recall, owing to selective episodic strengthening of their associations during study.     

Part-set cuing of false memories

Part-set cuing inhibition describes the common finding that re-presenting items from a word list can reduce subjects’ overall recall performance for studied items. Do part-set cuing effects occur for false memories as well? In the present experiments, subjects studied lists of words drawn from Roediger and McDermott (1995). After studying each list, subjects completed math problems and then recalled the list items either with or without accompanying list cues. In Experiment 1, the recall cues consisted of items drawn randomly from the original list. In Experiment 2, an additional type of cued recall task was added in which the even numbered list items were used as cues. Taken together, these experiments demonstrate robust part-set cuing effects for critical nonpresented items. In addition, they show that whereas recall of critical words is reduced by the presence of cues at test, retrieval cues do not affect critical words and studied words in exactly the same manner.     

A developmental study of learning disabilities and memory

Seven-, ten-, and thirteen-year-old learning-disabled (LD) and non-learning-disabled (NORM) children were presented specially structured lists of 38 words each and tested for free recall. Each list contained only four semantically related words. Two of the four related words were presented contiguously (serial positions 9 and 10) and the other two words were spaced (serial positions 20 and 30). All children recalled disproportionately more adjacent words (item 9 or 10) than any other words. Spaced words (items 20 and 30) were less likely to be recalled by the younger children than by the older children and by the LDs than by the NORMS. These findings provided support for the distinction between automatic and purposive semantic processing. NORMs' recall was governed by purposive semantic processing to a greater extent than was LDs' recall. However, no group or age differences were observed in automatic semantic processing.     

Comparison of intrusion errors and serial position curves on monaural and dichotic listening tasks: A developmental analysis

Based on a repeated measurement study of 7-, 10-, 14-, and 17-year-olds with monaural and dichotic presentations of word lists varying in associative structure and presentation rate, intrusion and serial position dependent variables were analyzed. Intrusion analyses supported previous reports on word list recall; vis, 7- and 10-year-olds recalled fewer words from monaural and dichotic lists and had more intrusive errors than adolescents. Based on similar orders of report (strategies) for monaural and dichotic word presentations, serial position curves for the two types of presentation were compared. Within-age comparisons were strikingly similar. Between-age comparisons of monaural and dichotic presentation serial position curves support the notion that there are ontogenetic limitations in memory structure and control processes.     

Effects of phonological similarity and concurrent irrelevant articulation on short-term-memory recall of repeated and novel word lists

The extent to which phonological similarity of list words impairs short-term-memory recall was investigated in two experiments. Experiment 1 showed that the phonological-similarity effect occurred both when list words were repeatedly sampled from a small set and when they were new on every trial, both when word-order information was required and when it was not. Furthermore, the adverse effect of phonological similarity on recall was apparent on the initial lists recalled, did not change over trials, and cannot be attributed to increasing levels of proactive inhibition across lists. In Experiment 2, subjects were required to count repeatedly to six during list presentation. Concurrent irrelevant articulation lowered recall and abolished the phonological similarity effect for both repeated and novel word lists.     

On the Meaning of Time in Later Life

Fourth-, sixth-, and eighth-grade subjects participated in a free-recall experiment to test the relative power of two Taxonomic, two Semantic Differential, and two Sense Impression word classes in facilitating recall. Findings show that across all grades children exhibited a significantly greater tendency to recall and cluster Taxonomic and Semantic Differential over Sense Impression word classes. Further, sixth and eighth graders recalled more words over all word classes than fourth graders. It was also found that girls recalled more words than boys across all three grade levels and all six word classes. Finally, there was no grade or sex effect for clustering. An interpretation of this difference was offered     

Transmitting false memories in social groups

In 2 experiments, subjects studied word lists drawn from Roediger and McDermott (1995) and then participated in perceived group recall (PGR) tests that were intended to lead each subject to believe that she or he was participating in collaborative recall in a 4-person group. Some of the lists were followed by PGR tests containing the nonpresented critical word, some lists were followed by PGR tests not containing the nonpresented critical word, and some lists were not followed by PGR tests. Subjects then completed individually administered recall tests and subsequent immediate or delayed recognition tests that required remember or know judgments. The major finding was that critical words contained within PGR tests were as likely to be falsely recalled, recognized, and consciously remembered as original list items. These findings show that false memories can be socially transmitted.     

The distinctiveness of the word-length effect

The authors report 2 experiments that compare the serial recall of pure lists of long words, pure lists of short words, and lists of long or short words containing just a single isolated word of a different length. In both experiments for pure lists, there was a substantial recall advantage for short words; the isolated words were recalled better than other words in the same list, and there was a reverse word-length effect: Isolated long words were recalled better than isolated short words. These results contradict models that seek to explain the word-length effect in terms of list-based accounts of rehearsal speed or in terms of item-based effects (such as difficulty of assembling items).     

Recall of short word lists presented visually at fast rates: Effects of phonological similarity and word length

Phonological similarity of visually presented list items impairs short-term serial recall. Lists of long words are also recalled less accurately than are lists of short words. These results have been attributed to phonological recoding and rehearsal. If subjects articulate irrelevant words during list presentation, both phonological similarity and word length effects are abolished. Experiments 1 and 2 examined effects of phonological similarity and recall instructions on recall of lists shown at fast rates (from one item per 0.114-0.50 sec), which might not permit phonological encoding and rehearsal. In Experiment 3, recall instructions and word length were manipulated using fast presentation rates. Both phonological similarity and word length effects were observed, and they were not dependent on recall instructions. Experiments 4 and 5 investigated the effects of irrelevant concurrent articulation on lists shown at fast rates. Both phonological similarity and word length effects were removed by concurrent articulation, as they were with slow presentation rates.     

Repetition deficits, list context, and word-class interactions in the RSVP of words in sentences.

We report a failure to find a repetition deficit in recall following the rapid serial visual presentation (RSVP) of words within sentences, using adjectives rather than nouns as the critical items. In a series of experiments that ruled out participant and procedural differences as the source of the failure, both word class and list context were found to moderate the repetition deficit, but grammatical necessity did not. The presence in the list of sentences in which the repeated adjectives were separated by more than three words (i.e., more than 400 ms in RSVP) not only eliminated the repetition deficit for the recall of those sentences but also for the recall of sentences in which the repeated adjectives were separated by three or fewer words (i.e., less than 400 ms in RSVP). However, although substantially reduced, a repetition deficit with noun-based materials was still found in this list context. Matching the adjective-based sentences with the noun-based sentences in sentence length and position of the critical items revealed that the moderating effect of word-class on the repetition deficit was mediated by the biases in sentence structure that using different word classes tend to induce.     

Generality of a strength model for three conditions of repeated recall

A Markovian model of repeated recall is presented that is based on four memory processes: (a) a presentation increment that strengthens the memory trace of a word when the word is shown to a subject, (b) a recall increment that strengthens the trace when the subject successfully recalls the word, (c) a cueing decrement that weakens the trace as contextual cues for retrieving a word change over successive trials of attempted recall, and (d) a sampling rule that translates the strength of a memory trace into a probability of recall. Mathematically, the model predicts the relative frequencies of the recall patterns that are defined by sequences of recalls and forgets for individual words over trials. According to the model, the recallability of a word depends on its state, which is a function of (a) whether the word has been recalled for the first time, and if so, then (b) how many consecutive forgets of the word immediately precede the current trial, (c) how many successful recalls of the word have accumulated, and (d) how many previous trials of attempted recall have elapsed. Paradigms that were well described by the model were (a) no reminding, in which words were presented once and then recalled repeatedly without additional presentations; (b) restricted reminding, in which words were presented before each trial of attempted recall up to their first successful recall but not thereafter; and (c) complete reminding, in which all words were presented befor each attempted recall, whether previously recalled or not. Differences in recall among these conditions were explained in terms of differences in values of the model's parameters.     

The fSAM model of false recall

The authors report a new theory of false memory building upon existing associative memory models and implemented in fSAM, the first fully specified quantitative model of false recall. Participants frequently intrude unstudied critical words while recalling lists comprising their strongest semantic associates but infrequently produce other extralist and prior-list intrusions. The authors developed the theory by simulating recall of such lists, using factorial combinations of semantic mechanisms operating at encoding, retrieval, or both stages. During encoding, unstudied words' associations to list context were strengthened in proportion to their strength of semantic association either to each studied word or to all co-rehearsed words. During retrieval, words received preference in proportion to their strength of semantic association to the most recently recalled single word or multiple words. The authors simulated all intrusion types and veridical recall for lists varying in semantic association strength among studied and critical words from the same and different lists. Multiplicative semantic encoding and retrieval mechanisms performed well in combination. Using such combined mechanisms, the authors also simulated several core findings from the Deese-Roediger-McDermott paradigm literature, including developmental patterns, specific list effects, association strength effects, and true-false correlations. These results challenge existing false-memory theories.     

Recall of morphologically complex forms is affected by memory task but not dyslexia

The authors studied the effect of morphological complexity on working memory in list recall tasks with base words (boy), inflected words (boy + 's) and derived words (boy + hood) in a morphologically rich language: Finnish. Simple serial recall was compared to complex working memory tasks, combining word recall with sentence verification in 8-year-old normally reading participants, dyslexic children, and adults. The normally reading children performed better than dyslexic children on both memory tasks and a test of morphology. Base words were better recalled than morphologically complex words. Memory was better for derived than inflected words in simple but not complex span tasks. There was no interaction between word type and reading group and thus no suggestion of dyslexia being associated with specific problems to represent complex morphology in working memory. Morphological processing in working memory appeared to depend on the task.     

False recognition and word length: A reanalysis of Roediger,Watson, McDermott,and Gallo (2001) and some new data

Roediger, Watson, McDermott, and Gallo (2001) reported a multiple regression analysis of the variables that predicted rates of false recall and recognition across lists in the Deese/Roediger— McDermott paradigm. They concluded that false recollection was predictable from the backward associative strength of critical words and list words and the veridical recall level of list words, with no independent contribution of frequency, concreteness, or length of critical words. A reanalysis of their data shows that critical word length does contribute to false recognition when it is measured relative to the length of other words in the list. Relative word length in the form of an unsignedz-score has a larger correlation with false recognition than any of the variables used by Roediger et al. and is also independently predictive of false recognition. This relationship was confirmed by the results of two recognition experiments in which false positives were significantly less frequent for words of lengths that never occurred in a studied list than for words of lengths that did occur in the study list.     

Independent effects of word concreteness and word valence on immediate serial recall

The present study examined the roles of word concreteness and word valence in the immediate serial recall task. Emotion words (e.g. happy) were used to investigate these effects. Participants completed study-test trials with 7-item study lists consisting of positive or negative words with either high or low concreteness (Experiments 1 and 2) and neutral (i.e. non-emotion) words with either high or low concreteness (Experiment 2). In serial recall performance, we replicated the typical item concreteness effect (concrete words are better recalled than abstract words) and obtained an item valence effect (positive/neutral words are better recalled than negative words). However, there was no concreteness × valence interaction. We conclude that both word valence and word concreteness independently contribute to the serial order retention of emotion words in the immediate serial recall task.     

Improving recall by recoding interfering material at the time of retrieval.

Our experiments demonstrate that interference of an interpolated list of items with recall of an original list can be substantially reduced by forming Ss just before testing how to reorganize and simplify the interpolated material. In Experiments 1 and 2, Ss better recalled an initial serial list of letters when informed at testing that an interpolated list spelled a certain phrase backward. Similarly, in Experiments 3 and 4, Ss better recalled an initial list of cities when told that the interpolated cities were also names of former U.S. presidents. Control experiments rule out several simple explanations. In contrast to an editing hypothesis, the postorganizing clue helped recall even when problems of list differentiation were minimized. Current memory models appear unable to explain this benefit of a postlearning clue that enables Ss to segregate the interpolated material from the to-be-remembered material.     

Redintegration and the benefits of long-term knowledge in verbal short-term memory: an evaluation of Schweickert's (1993) multinomial processing tree model

The impact of four long-term knowledge variables on serial recall accuracy was investigated. Serial recall was tested for high and low frequency words and high and low phonotactic frequency nonwords in 2 groups: monolingual English speakers and French-English bilinguals. For both groups the recall advantage for words over nonwords reflected more fully correct recalls with fewer recall attempts that consisted of fragments of the target memory items (one or two of the three target phonemes recalled correctly); completely incorrect recalls were equivalent for the 2 list types. However, word frequency (for both groups), nonword phonotactic frequency (for the monolingual group), and language familiarity all influenced the proportions of completely incorrect recalls that were made. These results are not consistent with the view that long-term knowledge influences on immediate recall accuracy can be exclusively attributed to a redintegration process of the type specified in multinomial processing tree model of immediate recall. The finding of a differential influence on completely incorrect recalls of these four long-term knowledge variables suggests instead that the beneficial effects of long-term knowledge on short-term recall accuracy are mediated by more than one mechanism.