Low-frequency Items Research Articles

Effects of word frequency on individual-item and serial order retention: Tests of the order-encoding view

The order-encoding view of the word frequency effect proposes that low-frequency (LF) items attract more attention to the encoding of individual-item information than do high-frequency (HF) items, but at the expense of order encoding (DeLosh & McDaniel, 1996). When combined with the assumption that free recall of unrelated words is organized according to their original order of presentation, this view explains the finding that HF words are better recalled than LF words in pure lists but that, in mixed lists, recall is better for LF words. The present study confirmed that in mixed lists, order memory becomes equivalent for HF and LF words and that the predicted pattern of order memory and recall holds fo r incidental order-encoding conditions, for longerlists than those used inprevious experiments, and for lists with minimal interitem associativity. Moreover, recall from HF lists declined, but recall from LF lists improved, in related-word lists, relative to unrelated-word lists, reversing the usual pure-list free recall advantage for HF words. These results were uniquely predicted by the order-encoding account and favor this view over accessibility, interitem association, and cuing effectiveness explanations of the word frequency effect.

Perception and recognition memory of words and werds: Two-way mirror effects

We examined associative priming of words (e.g., TOAD) and pseudohomophones of those words (e.g., TODE) in lexical decision. In addition to word frequency effects, reliable base-word frequency effects were observed for pseudohomophones: Those based on high-frequency words elicited faster and more accurate correct rejections. Associative priming had disparate effects on high- and low-frequency items. Whereas priming improved performance to high-frequency pseudohomophones, it impaired performance to low-frequency pseudohomophones. The results suggested a resonance process, wherein phonologic identity and semantic priming combine to undermine the veridical perception of infrequent items. We tested this hypothesis in another experiment by administering a surprise recognition memory test after lexical decision. When asked to identify words that were spelled correctly during lexical decision, the participants often misremembered pseudohomophones as correctly spelled items. Patterns of false memory, however, were jointly affected by base-word frequencies and their original responses during lexical decision. Taken together, the results are consistent with resonance accounts of word recognition, wherein bottom-up and top-down information sources coalesce into correct, and sometimes illusory, perception. The results are also consistent with a recent lexical decision model, REM-LD, that emphasizes memory retrieval and top-down matching processes in lexical decision.

The impact of phonological or semantic impairment on delayed auditory repetition: Evidence from stroke aphasia and semantic dementia

Background/Aims: This study aimed to evaluate the interactive account of repetition by examining the influence of factors that differentially tapped semantic and phonological processing in a case series of patients with semantic or phonological impairment.Methods & Procedures: We compared two patient groups: predominantly phonologically impaired cases with aphasia following cerebrovascular accident, and patients with semantic dementia. Immediate repetition was contrasted with repetition after a 5-second filled delay, and lexicality, frequency, and imageability were manipulated—therefore both the task and the neuropsychological impairment biased processing in favour of either lexical-semantic or phonological capacities.Outcomes & Results: Substantial interactivity was observed between phonological/semantic impairment and variables largely tapping these processes. The phonologically impaired patients showed substantial effects of lexicality and imageability that were larger in delayed than immediate repetition. The semantically impaired patients exhibited the complementary pattern, showing reduced effects of these lexical-semantic variables and a delay effect that was larger for more poorly comprehended, low-frequency items. Semantic errors were related to phonological deficits whereas semantic impairment led to an increase in phonological errors. The phonologically impaired stroke cases also made more perseverative responses.Conclusions: These findings support the view that repetition is underpinned by interaction between semantics and phonology within a single route and not by distinct lexical and sub-lexical pathways. The results also provide evidence of a continuum between phonological and deep dysphasia.

Data streaming algorithms for estimating entropy of network traffic

Using entropy of traffic distributions has been shown to aid a wide variety of network monitoring applications such as anomaly detection, clustering to reveal interesting patterns, and traffic classification. However, realizing this potential benefit in practice requires accurate algorithms that can operate on high-speed links, with low CPU and memory requirements. In this paper, we investigate the problem of estimating the entropy in a streaming computation model. We give lower bounds for this problem, showing that neither approximation nor randomization alone will let us compute the entropy efficiently. We present two algorithms for randomly approximating the entropy in a time and space efficient manner, applicable for use on very high speed (greater than OC-48) links. The first algorithm for entropy estimation is inspired by the structural similarity with the seminal work of Alon et al. for estimating frequency moments, and we provide strong theoretical guarantees on the error and resource usage. Our second algorithm utilizes the observation that the performance of the streaming algorithm can be enhanced by separating the high-frequency items (or elephants) from the low-frequency items (or mice). We evaluate our algorithms on traffic traces from different deployment scenarios.

The locus of naming difficulties in children with dyslexia: Evidence of inefficient phonological encoding

Twenty-four children with dyslexia (aged 7;7 to 12;1) and twenty-four age-matched controls named pictures aloud while hearing nonsense syllables either phonologically related (i.e., part of) or unrelated to the target picture name. Compared with controls, dyslexics had slower reaction times overall and, for low frequency items, the degree of facilitation from phonologically related sound segments relative to unrelated segments was proportionally greater. Within the dyslexic group, phonological facilitation was greater and picture naming speed slower for poorer compared with better readers. Phonological facilitation and picture naming speed was unrelated to reading ability in the normal readers. The results suggest that phonological encoding during speech production is less efficient in some children with dyslexia and that hearing part of the target word strongly facilitates this process. The findings are consistent with the hypothesis that dyslexia can arise from poorly specified output phonological representations.

“Presemantic” Cognition in Semantic Dementia: Six Deficits in Search of an Explanation

Abstract On the basis of a theory about the role of semantic knowledge in the recognition and production of familiar words and objects, we predicted that patients with semantic dementia would reveal a specific pattern of impairment on six different tasks typically considered “pre-” or “non-” semantic: reading aloud, writing to dictation, inflecting verbs, lexical decision, object decision, and delayed copy drawing. The prediction was that all tasks would reveal a frequency-by-typicality interaction, with patients performing especially poorly on lower-frequency items with atypical structure (e.g., words with an atypical spelling-to-sound relationship; objects with an atypical feature for their class, such as the hump on a camel, etc). Of 84 critical observations (14 patients performing 6 tasks), this prediction was correct in 84/84 cases; and a single component in a factor analysis accounted for 87% of the variance across seven measures: each patient's degree of impairment on atypical items in the six experimental tasks and a separate composite score reflecting his or her degree of semantic impairment. Errors also consistently conformed to the predicted pattern for both expressive and receptive tasks, with responses reflecting residual knowledge about the typical surface structure of each domain. We argue that these results cannot be explained as associated but unrelated deficits but instead are a principled consequence of a primary semantic impairment.

"Presemantic" Cognition in Semantic Dementia: Six Deficits in Search of an Explanation

On the basis of a theory about the role of semantic knowledge in the recognition and production of familiar words and objects, we predicted that patients with semantic dementia would reveal a specific pattern of impairment on six different tasks typically considered "pre-" or "non-" semantic: reading aloud, writing to dictation, inflecting verbs, lexical decision, object decision, and delayed copy drawing. The prediction was that all tasks would reveal a frequency-by-typicality interaction, with patients performing especially poorly on lower-frequency items with atypical structure (e.g., words with an atypical spelling-to-sound relationship; objects with an atypical feature for their class, such as the hump on a camel, etc). Of 84 critical observations (14 patients performing 6 tasks), this prediction was correct in 84/84 cases; and a single component in a factor analysis accounted for 87% of the variance across seven measures: each patient's degree of impairment on atypical items in the six experimental tasks and a separate composite score reflecting his or her degree of semantic impairment. Errors also consistently conformed to the predicted pattern for both expressive and receptive tasks, with responses reflecting residual knowledge about the typical surface structure of each domain. We argue that these results cannot be explained as associated but unrelated deficits but instead are a principled consequence of a primary semantic impairment.

The variance theory of the mirror effect in recognition memory.

The mirror effect refers to a rather general empirical finding showing that, for two classes of stimuli, the class with the higher hit rates also has a lower false alarm rate. In this article, a parsimonious theory is proposed to account for the mirror effect regarding, specifically, high- and low-frequency items and the associated receiver-operating curves. The theory is implemented in a recurrent network in which one layer represents items and the other represents contexts. It is shown that the frequency mirror effect is found in this simple network if the decision is based on counting the number of active nodes in such a way that performance is optimal or near optimal. The optimal performance requires that the number of active nodes is low, only nodes active in the encoded representation are counted, the activation threshold is set between the old and the new distributions, and normalization is based on the variance of the input. Owing to the interference caused by encoding the to-be-recognized item in several preexperimental contexts, the variance of the input to the context layer is greater for high- than for low-frequency items, which yields lower hit rates and higher false alarm rates for high- than for low-frequency items. Although initially the theory was proposed to account for the mirror effect with respect to word frequency, subsequent simulations have shown that the theory also accounts for strength-based mirror effects within a list and between lists. In this case, consistent with experimental data, the variance theory suggests that focusing attention to the more difficult class within a list affects the hit rate, but not the false alarm rate and not the standard deviations of the underlying density, leading to no mirror effect.

Patterns of naming objects and actions in children with word finding difficulties

Children who experience difficulties in naming are described as having word finding difficulties (WFDs). In the present study 31 children with WFDs were identified through a wider survey of educational provision for those with language and communication difficulties. The children were included if they were between 6;4-7;10 years, had normal non-verbal intelligence, no major articulation difficulties, and had WFDs as diagnosed by the Test of Word Finding Difficulties. Three control groups were identified who were matched on: chronological age (N = 31), naming age (N = 31), and level of receptive grammar (N = 31). Children's comprehension of words, accuracy of naming, and latency to name were assessed for numerals, letters, and pictures of objects and actions. Half the pictures presented were high frequency items and half were low frequency items. The children with WFDs formed a heterogeneous group with respect to other language measures with the primary defining feature being their poor performance in word retrieval and their poor performance on a semantic fluency test. No differences were found between the children with WFDs and their age-matched peers when naming letters and numerals or in the comprehension of objects and actions. In contrast, the accuracy and latency of naming were significantly worse than that of age-matched peers. Accuracy of naming was equivalent to that of the language-matched peers and error patterns also were similar. However, the children with WFDs were the slowest to provide responses and for naming high frequency objects this difference approached significance in comparison to language-matched peers (p = .052). The findings point to the importance of using appropriate control groups, and are discussed in relation to the idea that WFDs are caused by impoverished semantic representations.

When Ottoman is Easier than Chair: An Inverse Frequency Effect in Jargon Aphasia

This paper presents evidence of an inverse frequency effect in jargon aphasia. The subject (JP) showed a pre-disposition for low frequency word production on a range of tasks, including picture naming, sentence completion and naming in categories. Her real word errors were also striking, in that these tended to be lower in frequency than the target. Reading data suggested that the inverse frequency effect was present only when production was semantically mediated. It was therefore hypothesised that the effect was at least partly due to the semantic characteristics of low frequency items. Some support for this was obtained from a comprehension task showing that JP's understanding of low frequency terms, which she often produced as errors, was superior to her understanding of high frequency terms. Possible explanations for these findings are considered.

The role of sub-lexical orthography in naming: a performance and acoustic analysis

Two experiments examined whether the duration of naming is sensitive to variables that influence pre-production processes. In Experiment 1 word naming produced longer durations than picture naming. Experiment 2 replicated this effect for low frequency items, but not for high frequency items. The results indicate that naming duration increases when conditions are likely to involve sub-lexical orthographic to phonological translation. The results are inconsistent with the proposition that input and output involve discrete or distinct information processing stages in naming where the whole response is planned in advance of response initiation. Rather, it is claimed that sub-lexical information about a printed word's pronunciation can be used to initiate a naming response before the whole response is fully prepared. The implications for models of word naming and the role of suprasegmental factors in naming are discussed.

Performance of elderly white and African American community residents on the abbreviated CERAD Boston naming test

Differences in the responses of an elderly biracial group of cognitively normal subjects to a 15-item short version of the Boston Naming Test developed for the Consortium to Establish a Registry for Alzheimer's Disease (CERAD) were examined. The subjects consisted of 103 Whites and 136 African Americans who were 70 years of age and older and living in a five-county urban and rural area of North Carolina. They were drawn from the Duke University site of the Established Populations for Epidemiologic Studies of the Elderly (EPESE). All were cognitively normal. With gender, years of education, and age controlled, White subjects performed significantly better than did African American subjects. The items in this test were selected to represent words with a high, medium, and low frequency of occurrence in English. They did not, however, show the expected gradation for either racial group. Medium and low frequency items were of comparable difficulty for the two races. Hierarchical ordering of difficulty would be improved with minor rearrangement of items.

Immediate serial recall, word frequency, item identity and item position.

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.

The structural organization of the mental lexicon and its contribution to age-related declines in spoken word recognition

Three experiments were conducted within the framework of the Neighborhood Activation Model of spoken word recognition to examine how the structural organization of the mental lexicon may contribute to age-related declines in spoken language processing. According to the model, lexically hard words are items that are phonetically similar to many other high-frequency words and should be identified less accurately than easy words which are phonetically similar to relatively few, low-frequency items. The results of experiment 1 demonstrated that lexical difficulty had differential effects on older and younger listeners. Specifically, relative to young listeners, older adults exhibited significantly poorer identification scores for lexically hard items. Experiment 2 indicated that these age differences in the effects of lexical difficulty were maintained under conditions in which performance for easy words was the same for older and younger listeners. In the third experiment, reducing the resources available for perceptual identification, by changing from single to multiple talkers, produced greater effects of lexical difficulty for older, compared to younger, listeners. Explanations for the results based on diminished cognitive resources and impaired inhibitory control will be discussed. [Work supported by Washington Univ.]

Event-Related Potentials and Recognition Memory for Low- and High-Frequency Words

Event-related potentials (ERPs) were recorded while subjects made recognition judgments on high- and low-frequency words, half of which had previously been presented in an incidental study task. Compared to high-frequency items, low-frequency words were associated with superior recognition performance, and attracted a higher proportion of confident judgments. In the case of the low-frequency words only, the region of the ERPs post-500 msec evoked by correctly classified, previously studied (old) words was more positive-going than was the same region of the EWs to nonstudied (new) words. These "old/new" ERP differences were larger from electrodes over the left than over the right hemisphere. This old/new by frequency interaction held when EWs were formed only from words that attracted confident judgments. It is argued that these data are consistent with the ideas that (1) post-500 msec "old/ new" EW differences in recognition memory tasks reflect differences in old and new words' levels of relative familiarity, and (2) the recognition memory advantage for low-frequency words results, at least in part, from the higher level of relative familiarity engendered at test by previously studied low-frequency items. The data are interpreted as providing support for "two-process" models of recognition memory.

The Interaction of Lexical and Non-Lexical Processing Mechanisms: Evidence from Anomia

Patient SF presented with a selective deficit in naming, in the presence of normal auditory, visual and pictorial input processing, and of normal comprehension. The naming disorder was independent of input modality, and resulted in the inability to respond to lowfrequency items. The analysis of the patient's performance in the repeated administration of the same set of pictures for oral naming and for written naming demonstrated that both tasks were disrupted to a similar extent - high consistency values were observed both across and within modality. Thus, the profile of the naming disorder observed in this patient was consistent with damage to phonological and orthographic output lexicons. SF's performance in reading and writing was apparently at odds with this account, as he was able to read aloud words with lexically-assigned stress, and to write to dictation words whose spelling is also determined lexically. The co-occurrence of the inability to activate output lexical representations in both oral and written picture naming, in the context of normal ability to activate the same representations in reading and in writing is consistent with the notion that phonological and orthographic output lexicons are distinct, but interact with nonlexical conversion mechanisms.

Degree of Cognitive Impairment and the Dissociation of Implicit and Explicit Memory

One group of normal elderly adults and two groups of cognitively impaired elderly adults were compared on an implicit and an explicit memory task. Degree of impairment affected explicit memory; the mildly and moderately impaired elderly adults demonstrated significantly reduced recognition performance. Degree of impairment also affected implicit memory priming performance, particularly for low frequency items, as measured by a word completion task. Explicit memory performance declined for both cognitively impaired groups, but implicit memory performance was sensitive to the degree or severity of impairment, causing a decline in performance only in the moderately impaired group.

Naming multisyllabic words.

The process of reading multisyllabic words aloud from print was examined in 4 experiments. Experiment 1 used multisyllabic words that vary in terms of the consistency of component spelling-sound correspondences. The stimuli were regular, regular inconsistent, and exception words analogous to the monosyllabic items used in previous studies. Both regular inconsistent and exception words produced longer naming latencies than regular words. In Experiment 2 these differences between word types were found to be limited to lower frequency items. Experiment 3 showed that effects of number of syllables on naming latency are also limited to lower frequency words when the stimulus display forced subjects to use syllabic units. Thus, frequency modulates the effects of two aspects of lexical structure-consistency of spelling-sound correspondences and number of syllables. The results suggest that the naming of multisyllabic words draws on some of the same knowledge representations and processes as monosyllabic words; however, naming does not require syllabic decomposition. The results are discussed in the context of current models of naming.

A subquadratic algorithm for constructing approximately optimal binary search trees

An algorithm is presented which constructs an optimal binary search tree for an ordered list of n items, and which requires subquadratic time if there is no long sublist of very low frequency items. For example, time = O( n 1.6) if the frequency of each item is at least ε n for some constant ε > 0. A second algorithm is presented which constructs an approximately optimal binary search tree. This algorithm has one parameter, and exhibits a trade-off between speed and accuracy. It is possible to choose the parameter such that time = O( n 1.6) and error = o(1).

Performance Evaluation of Occupational Therapy Students: A Validity Study

Objective evaluation of clinical competence is always a difficult task. Many attempts have been made to design an instrument for such evaluation. and Performance Evaluation of Occupational Therapy Students (PEOTS) is such an instrument. It was designed as a criterion-referenced evaluation of performance in four broad areas of clinical competence: professional interaction, assessment, programming and treatment. This instrument has enjoyed fairly broad acceptance based on its face validity alone. The purpose of this study was to collect preliminary data on the content and construct validity of the instrument. A sample of 43 new graduates from Canadian occupational therapy programs were evaluated on the job within the first year of their practice using PEOTS. The data were collected over a two year period and came from facilities across the country. Analysis of the results showed that the instrument had overall good content and construct validity. However, some low frequency items need to be investigated further and some items need to have their criterion levels changed.