Fernando Cuetos

Cross-linguistic differences in parsing: Restrictions on the use of the Late Closure strategy in Spanish ☆

Abstract Two questionnaire studies and three on-line experiments were carried out in an attempt to determine whether a parsing strategy which is widely used in English (the Late Closure strategy) simply represents an arbitrarily chosen procedure or whether it is likely to have become a relatively stable feature of the parsing routine because it confers real information processing advantages on the reader or listener. Experiment 1A showed that Spanish readers apparently do not use the Late Closure strategy to interpret a particular type of sentence. With comparable materials, however, English readers do use this strategy (Experiment 1B). Experiments 2, 3 and 4 showed that the structural choices were made while the subjects were actually reading the sentence. The results throw doubt on the suggestion that the Late Closure strategy is favoured (in English) mainly because it is efficient in information processing terms. It may be that different languages make use of parsing strategies in an essentially arbitrary way. Implications for general theories of human parsing are discussed.

Exposure-based models of human parsing: Evidence for the use of coarse-grained (nonlexical) statistical records

Several current models of human parsing maintain that initial structural decisions are influenced (or tuned) by the listeners or readers prior contact with language. The precise workings of these models depend upon the “grain,” or level of detail, at which previous exposures to language are analyzed and used to influence parsing decisions. Some models are premised upon the use of fine-grained records (such as lexical cooccurrence statistics). Others use coarser measures. The present paper considers the viability of models based exclusively on the use of fine-grained lexical records. The results of several studies are reviewed and the evidence suggests that, if they are to account for the data, experience-based parsers must draw upon records or representations that capture statistical regularities beyond the lexical level. This poses problems for several parsing models in the literature.

Naming times for the Snodgrass and Vanderwart pictures in Spanish.

We present new Spanish norms for object familiarity and rated age of acquisition for 140 pictures taken from Snodgrass and Vanderwart (1980), together with data on visual complexity, image agreement, name agreement, word length (in syllables and phonemes), and five measures of word frequency. The pictures were presented to a group of 64 Spanish subjects, and oral naming latencies were recorded. In a multiple regression analysis, age of acquisition, object familiarity, name agreement, word frequency, and word length made significant independent contributions to predicting naming latency.

Aphasic naming in Spanish: predictors and errors.

Sixteen Spanish aphasic patients named drawings of objects on three occasions. Multiple regression analyses were carried out on the naming accuracy scores. For the patient group as a whole, naming was affected by visual complexity, object familiarity, age of acquisition, and word frequency. The combination of variables predicted naming accuracy in 15 of the 16 individual patients. Age of acquisition, word frequency, and object familiarity predicted performance in the greatest number of patients, while visual complexity, imageability, animacy, and length all affected performance in at least two patients. High proportions of semantic and phonological errors to particular objects were associated with objects having early learned names while high proportions of no-response errors were associated with low familiarity and low visual complexity. It is suggested that visual complexity and object familiarity affect the ease of object recognition while word frequency affects name retrieval. Age of acquisition may affect both stages, accounting for its influence in patients with a range of different patterns of disorder.

Action naming is impaired in Parkinson disease patients

In order to explore the possible contribution of the motor system to the representation of verbs, we studied the relative preservation of the capacity of Parkinson disease patients to name matched sets of object and action pictures. The performance of this group of participants was compared with that of a group of healthy seniors, and a group of Alzheimer disease patients. Generalized linear mixed-effects analyses showed that, whereas the two control groups had similar accuracy scores in response to objects and actions, Parkinson disease patients presented a significant impairment in their capacity to name actions compared to objects. The results of this study support the idea that verb representations are grounded in neural networks to which brain areas involved in motor control contribute.

Word naming in Spanish

Studies on the factors that determine word naming speed have been conducted in a number of languages. In this study two objectives were pursued: (1) to contrast the effects of the variable of age of acquisition with the different measures of frequency of use (adult written frequency, child written frequency, cumulative frequency, and frequency trajectory), and (2) to verify which variables determine reading latencies in a language with a completely transparent reading system (Spanish). For these purposes, 53 native speakers of Spanish read aloud 240 words for which values were available of the main lexical, phonological, and semantic variables. The results of the regression analyses showed that the best predictors of reading times were length (measured in terms of both number of letters and number of syllables) and subjective age of acquisition (AoA). The different frequency measures correlated significantly with the response latencies but were not longer significant in the regression analyses. Semantic variables (familiarity and imageability) had the smallest effects on reading speed. The comparison of the results obtained in this study with those reported for other languages offers some conclusions about the effects of orthographic systems on word naming.

Writing Processes in a Shallow Orthography.

The aim of this study was to find out if the spelling models proposed to account for performance in languages with a deep orthography are applicable to a shallow orthography such as Spanish. For this reason, two experiments were carried out, one of lexical decision (in which subjects heard words and nonwords but only wrote down the nonwords) and the other a dictation in which the variables of grapheme frequency and lexical priming were manipulated. The results coincide with those found in the English language as in both experiments the two variables produced significant effects: the variable of grapheme frequency, which indicates the existence of a process of phonological spelling and the variable of lexical priming which indicates the existence of a lexical process.

Linguistic changes in verbal expression: A preclinical marker of Alzheimer's disease

Despite the many studies examining linguistic deterioration in Alzheimers disease (AD), very little is known about changes in verbal expression during the preclinical phase of this disease. The objective of this study was to determine whether changes in verbal expression occur in the preclinical phase of AD. The sample consisted of 40 healthy Spanish speakers from Antioquia, Colombia. A total of 19 were carriers of the E280A mutation in the Presenilin 1 gene, and 21 were noncarrier family members. The two groups were similar in age and education. All the participants were shown the Cookie Theft Picture Card from the Boston Diagnostic Aphasia Examination and were asked to describe the scene. Specific grammatical and semantic variables were evaluated. The performance of each group was compared using multivariate analyses of the variance for semantic and grammatical variables, and errors. Carriers of the mutation produced fewer semantic categories than noncarriers. In the preclinical phase of AD, changes in verbal expression are apparent and early detection of these differences may assist the early diagnosis of and intervention in this disease.

Normative data and naming times for action pictures

The present article provides Spanish norms for name agreement, printed word frequency, word compound frequency, familiarity, imageability, visual complexity, age of acquisition, and word length (measured by syllables and phonemes) for 100 line drawings of actions taken from Druks and Masterson (2000). In addition, through a naming-time experiment carried out with a group of 54 Spanish students in a pool of 63 of these line drawings, we determined the best predictors of naming actions. In the multiple regression analysis, age of acquisition and name agreement emerged as the most important determinants of action-naming reaction time.

Neural correlates of abstract verb processing

The present study investigated the neural correlates of the processing of abstract (low imageability) verbs. An extensive body of literature has investigated concrete versus abstract nouns but little is known about how abstract verbs are processed. Spanish abstract verbs including emotion verbs (e.g., amar, “to love”; molestar, “to annoy”) were compared to concrete verbs (e.g., llevar, “to carry”; arrastrar, “to drag”). Results indicated that abstract verbs elicited stronger activity in regions previously associated with semantic retrieval such as inferior frontal, anterior temporal, and posterior temporal regions, and that concrete and abstract activation networks (compared to that of pseudoverbs) were partially distinct, with concrete verbs eliciting more posterior activity in these regions. In contrast to previous studies investigating nouns, verbs strongly engage both left and right inferior frontal gyri, suggesting, as previously found, that right prefrontal cortex aids difficult semantic retrieval. Together with previous evidence demonstrating nonverbal conceptual roles for the active regions as well as experiential content for abstract word meanings, our results suggest that abstract verbs impose greater demands on semantic retrieval or property integration, and are less consistent with the view that abstract words recruit left-lateralized regions because they activate verbal codes or context, as claimed by proponents of the dual-code theory. Moreover, our results are consistent with distributed accounts of semantic memory because distributed networks may coexist with varying retrieval demands.