Randi C. Martin

How semantic is automatic semantic priming

Priming for semantically related concepts was investigated using a lexical decision task designed to reveal automatic semantic priming. Two experiments provided further evidence that priming in a single presentation lexical decision task (McNamara & Altarriba, 1988) derives from automatic processes. Mediated priming, but no inhibition or backward priming was found in this type of lexical decision task. Experiments 3 and 4 demonstrated that automatic priming was found only for associated word pairs, as determined by word association norms, and not for word pairs that are semantically related but not associated. It is argued that automatic priming in the lexical decision task occurs at a lexical level not at a semantic level.

Semantic and phonological codes interact in single word production.

The relationship between semantic-syntactic and phonological levels in speaking was investigated using a picture naming procedure with simultaneously presented visual or auditory distractor words. Previous results with auditory distractors have been used to support the independent stage model (e.g., H. Schriefers, A. S. Meyer, & W. J. M. Levelt, 1990), whereas results with visual distractors have been used to support an interactive view (e.g., P.A. Starreveld & W. La Heij, 1996b). Experiment 1 demonstrated that with auditory distractors, semantic effects preceded phonological effects, whereas the reverse pattern held for visual distractors. Experiment 2 indicated that the results for visual distractors followed the auditory pattern when distractor presentation time was limited. Experiment 3 demonstrated an interaction between phonological and semantic relatedness of distractors for auditory presentation, supporting an interactive account of lexical access in speaking.

Verbal working memory and sentence comprehension: A multiple-components view.

Semantic and syntactic aspects of sentence comprehension were investigated for 3 patients who showed different patterns of performance on short-term memory tasks. On a sentence anomaly judgment task assessing the retention of semantic information, only patient A.B. showed a detrimental effect on comprehension with increases in the number of words to be held in an unintegrated fashion. On judgments of grammatical acceptability, only patient M.W. demonstrated a detrimental effect of increasing the number of words intervening between the words signaling that a sentence was ungrammatical. The results suggest that semantic and syntactic components must be postulated in addition to the phonological and articulatory components of A. D. Baddeleys (1986, 1990) working memory model. Although the phonological, semantic, and syntactic components may be differentiall y affected by brain damage, the components interact and support each other in normal comprehension. The working memory model initially proposed by Baddeley and Hitch (1974) has played an important role in neuropsychological studies of brain-damaged patients with language disorders. The essential insight of their model was that working memory should not be seen as a unitary store, but rather as consisting of different components with separate storage capacities. It differed in this respect from the so-called modal model of memory (Murdock, 1974), which posited a single store in which information of all types was maintained prior to transfer to long-term memory (e.g., Atkinson & Shiffrin, 1968; Waugh & Norman, 1965). Baddeley and Hitchs now well-known model separated a central executive component of working memory from peripheral storage devices dedicated to the maintenance of different types of information. The central executive was seen as an attention-controlling system with storage space of its own, which directed the processes involved in cognitive functions and which drew on the information stored in the peripheral stores. The peripheral storage devices included a phonemically based rehearsal buffer, currently termed the phonological loop, and a visual buffer, currently termed the visuospatial sketchpad (Baddeley, 1986, 1990). The phonological loop itself may be broken down into two components, a phonological storage component and an articulatory rehearsal component (Salame & Baddeley, 1982; Vallar & Baddeley, 1984a).

Reading comprehension in the presence of unattended speech and music

This series of experiments investigated whether the detrimental effects of unattended speech that have been obtained in short-term memory tasks would be obtained in reading comprehension. Such effects would be expected if reading comprehension depends on the maintenance of phonological information in short-term memory. The first three experiments demonstrated that unattended speech but not music interfered with reading comprehension while unattended music had a greater interfering effect than speech on a music identification task. Experiments 4 and 5 showed that the detrimental effect of the speech backgrounds on reading was due to their semantic rather than their phonological properties. The failure to find a phonological interference effect argues against a role for phonological short-term memory in reading comprehension.

Dissociations among tasks involving inhibition: A single-case study

Recent theories of working memory have emphasized the role of inhibition in suppressing irrelevant information. Moreover, psychometric studies have reported that several inhibition tasks with very diverse requirements load on a single inhibition factor. A patient with left inferior frontal damage, Patient M.L., previously reported to have a semantic short-term memory deficit (R. C. Martin & He, 2004), showed evidence of difficulty with inhibition on short-term memory tasks. We investigated whether he would show evidence of inhibition difficulty on two verbal tasks (a Stroop task and a recent-negatives task) and two nonverbal tasks (a nonverbal spatial Stroop task and an antisaccade task). M.L. was impaired on both verbal tasks but performed normally on the nonverbal tasks. M.L.’s data also represent a dissociation between Stroop and antisaccade performance, two tasks that load on a single factor in factor-analytic studies. The implications of these data for theories of inhibition and executive function are discussed.

Gender and letters of recommendation for academia: agentic and communal differences.

In 2 studies that draw from the social role theory of sex differences (A. H. Eagly, W. Wood, & A. B. Diekman, 2000), the authors investigated differences in agentic and communal characteristics in letters of recommendation for men and women for academic positions and whether such differences influenced selection decisions in academia. The results supported the hypotheses, indicating (a) that women were described as more communal and less agentic than men (Study 1) and (b) that communal characteristics have a negative relationship with hiring decisions in academia that are based on letters of recommendation (Study 2). Such results are particularly important because letters of recommendation continue to be heavily weighted and commonly used selection tools (R. D. Arvey & T. E. Campion, 1982; R. M. Guion, 1998), particularly in academia (E. P. Sheehan, T. M. McDevitt, & H. C. Ross, 1998).

Short-term memory and sentence processing: Evidence from neuropsychology

Traditional models of memory assume that short-term memory, as measured by memory span, plays an important role in linguistic processing and the learning ofverbal information. Contradicting this view are findings from a brain-damaged patient, E.A., who, despite a verbal memory span of about two items, demonstrated normal sentence comprehension in a variety oftasks. She was, however, impaired whenever verbatim phonological information had to be maintained or learned. These results and those from other patients with reduced span suggest that the phonological storage capacity that is critical to memory span plays only a limited role in language processing, specifically in the maintenance and learning of phonological forms. Implications for models of short-term memory are discussed. It is argued that short-termmemory should be seen as deriving from the processing and retentive capacities of language processing modules, with span tasks drawing on only a subset of these modules.

Executive function in older adults: A structural equation modeling approach.

Confirmatory factor analysis (CFA) and structural equation modeling (SEM) were used to study the organization of executive functions in older adults. The four primary goals were to examine (a) whether executive functions were supported by one versus multiple underlying factors, (b) which underlying skill(s) predicted performance on complex executive function tasks, (c) whether performance on analogous verbal and nonverbal tasks was supported by separable underlying skills, and (d) how patterns of performance generally compared with those of young adults. A sample of 100 older adults completed 10 tasks, each designed to engage one of three control processes: mental set shifting (Shifting), information updating or monitoring (Updating), and inhibition of prepotent responses (Inhibition). CFA identified robust Shifting and Updating factors, but the Inhibition factor failed to emerge, and there was no evidence for verbal and nonverbal factors. SEM showed that Updating was the best predictor of performance on each of the complex tasks the authors assessed (the Tower of Hanoi and the Wisconsin Card Sort). Results are discussed in terms of insight for theories of cognitive aging and executive function.

Semantic short-term memory and its role in sentence processing: A replication ☆

Previous studies have shown that an aphasic patient (AB) with a semantic short-term memory deficit (STM) had difficulties comprehending and producing sentences with structures that demanded the simultaneous retention of several individual word meanings (Martin & Freedman, 2001a, 2001b; Martin & Romani, 1994; Martin, Shelton, & Yaffee, 1994). The present study provides a replication of these findings with an additional case (ML) who shows a striking dissociation between preserved semantic knowledge and disrupted semantic STM. ML performed poorly on comprehension tasks for sentences that required the retention of single word meanings across several intervening words. In contrast, he did not show an effect of intervening words on processing grammatical relations. ML had difficulty producing adjective-noun phrases, though able to produce the individual nouns and adjectives. These findings support the contention that there is a semantic retention capacity, involved in both comprehension and production, that is separate from semantic knowledge representations.

Dissociable components of short-term memory and their relation to long-term learning.

Short-term memory (STM) includes dissociable phonological and semantic components (R.C. Martin, 1993). Previous findings indicate that phonological STM capacity supports learning of novel phonological forms, such as new vocabulary (e.g., Baddeley, Gathercole, & Papagno, 1998). It was hypothesised that semantic STM capacity would support the learning of novel semantic information. Five aphasic patients were tested who demonstrated deficits in the short-term retention of either phonological or semantic information. Four of the five patients demonstrated learning deficits in a paired associate paradigm that corresponded to their STM deficits. One patient with a severe deficit in phonological STM but a better-preserved ability to retain semantic information showed better learning of new semantic information than new phonological information. Three patients with a greater deficit in semantic than phonological STM showed the reverse. A fifth patient with a severe semantic STM deficit failed to show learning for either type of material. Results suggest that the semantic and phonological components of STM are essential for the long-term learning of corresponding representations in long-term memory.