Orna Peleg

Salience and Context Effects: Two Are Better Than One

This study provides evidence supporting the hypothesis that language comprehension involves 2 separate mechanisms that run in parallel: a linguistic mechanism and a contextual mechanism. The linguistic mechanism (e.g., lexical access) is modular and stimulus driven; it is a bottom-up, perceptual mechanism, induced by a lexical stimulus to search the mental lexicon for its match. This mechanism is encapsulated with respect to nonlinguistic information and thus operates locally (i.e., on the word level). Lexical access is exhaustive and ordered: Salient meanings are accessed faster. Contextual facilitation, on the other hand, is the outcome of a central, expectation-driven mechanism that operates globally during language comprehension at the point where prior linguistic information has already been processed and interfaced with other cognitive processes (e.g., inferencing). Experiment 1 indicates that contextual facilitation can occur even before lexical accessing takes place, fostering an impression of a selective process. Experiment 2 shows that the target words position in the sentence (initial vs. noninitial) is crucial for the operation of the predictive mechanism. Thus, we would not expect contextual meanings to outweigh salient meanings at the beginning of sentences.

Hemispheric Sensitivities to Lexical and Contextual Information: Evidence from Lexical Ambiguity Resolution.

The present study examined the manner in which both hemispheres utilize prior semantic context and relative meaning frequency during the processing of homographs. Participants read sentences biased toward the dominant or the subordinate meaning of their final homograph, or unbiased neutral sentences, and performed a lexical decision task on lateralized targets presented 250ms after the onset of the sentence-final ambiguous prime. Targets were either related to the dominant or the subordinate meaning of the preceding homograph, or unrelated to it. Performance asymmetry was found in the absence of a biasing context: dominant-related targets were exclusively facilitated in the RVF/LH, whereas both dominant- and subordinate-related targets were facilitated in the LVF/RH. Performance symmetry was found in the presence of a biasing context: dominant-related targets were exclusively activated in dominant-biasing contexts, whereas both dominant- and subordinate-related targets were facilitated in subordinate-biasing contexts. The implications of the results for both general and hemispheric models of word processing are discussed.

Semantic asymmetries are modulated by phonological asymmetries : Evidence from the disambiguation of homophonic versus heterophonic homographs

The present study investigated cerebral asymmetries in accessing multiple meanings of two types of homographs: homophonic homographs (e.g., bank) and heterophonic homographs (e.g., tear). Participants read homographs preceded by either a biasing or a non-biasing sentential context and performed a lexical decision on lateralized targets presented 150 ms after onset of the sentence-final ambiguous prime. Targets were either related to the dominant or the subordinate meaning of the preceding homograph or were unrelated to it. In the case of homophonic homographs--our results converge with previous findings: both activation and selection processes are faster in the LH than in the RH. Importantly, however, in the case of heterophonic homographs--opposite asymmetries were found. These results suggest that semantic asymmetries are modulated by phonology. They are discussed in the context of a model of functional architecture of reading in the two hemispheres in which orthography, phonology and semantics are fully interconnected in the LH, whereas in the RH, orthography and phonology are not directly connected, such that phonological processes are mediated by semantics.

Contextual Strength: the Whens and Hows of Context Effects

Highlighting the role context plays in shaping our linguistic behaviour is the major contribution of pragmatics to language research. Indeed, pragmatics has shifted the focus of research from the code to contextual inference (Carston, 2002; Sperber & Wilson, 1986/1995). It is widely agreed now that contextual information is a crucial factor determining how we make sense of utterances. The role of context is even more pronounced within a framework that assumes that the code is underspecified allowing for top-down inferential processes to narrow meanings down and adjust them to the specific context.

Hemispheric asymmetries in meaning selection: Evidence from the disambiguation of homophonic vs. heterophonic homographs

Research investigating hemispheric asymmetries in meaning selection using homophonic homographs (e.g., bank), suggests that the left hemisphere (LH) quickly selects contextually relevant meanings, whereas the right hemisphere (RH) maintains a broader spectrum of meanings including those that are contextually irrelevant (e.g., Faust & Chiarello, 1998). The present study investigated cerebral asymmetries in maintaining the multiple meanings of two types of Hebrew homographs: homophonic homographs and heterophonic homographs (e.g., tear). Participants read homographs preceded by a biasing, or a non-biasing sentential context, and performed a lexical decision task on targets presented laterally, 1000ms after the onset of the sentence-final ambiguous prime. Targets were related to either the dominant or the subordinate meaning of the preceding homograph, or unrelated to it. When targets were presented in the LVF/RH, dominant and subordinate meanings, of both types of homographs, were retained only when they were supported by context. In a non-biasing context, only dominant meanings of homophonic homographs were retained. Alternatively, when targets were presented in the RVF/LH, priming effects for homophonic homographs were only evident when meanings were supported by both context and frequency (i.e., when context favored the dominant meaning). In contrast, heterophonic homographs resulted in activation of dominant meanings, in all contexts, and activation of subordinate meanings, only in subordinate-biasing contexts. The results challenge the view that a broader spectrum of meanings is maintained in the right than in the left hemisphere and suggest that hemispheric differences in the time course of meaning selection (or decay) may be modulated by phonology.

Controlled semantic processes within and between the two cerebral hemispheres

ABSTRACT To test the separate and combined abilities of the two cerebral hemispheres to perform controlled semantic selection and integration processes, Hebrew readers saw pairs of words and had to decide whether the two words were semantically related. The first word in each pair was presented centrally. The second word was presented in the left, right, or central visual field (LVF, RVF, and CVF). We compared response latencies for related pairs in two conditions: In the ambiguous condition, the first word was a homograph (either homophonic or heterophonic) and the second word was related to either its dominant or subordinate meaning. In the unambiguous condition, homographs were replaced with unambiguous control words. Irrespective of VF or homograph type, response times for ambiguous pairs were significantly longer than for unambiguous pairs only when targets were related to the subordinate meaning of the homograph. In the left hemisphere (RVF/LH), this ambiguity effect was larger for heterophones than for homophones, whereas in the right hemisphere (LVF/RH), similar patterns were observed for both types of homographs. Finally, performance patterns in the CVF revealed the same patterns as those in the RVF/LH, and were different from those in the LVF/RH. The implications of these results are discussed.

Transfer of L1 visual word recognition strategies during early stages of L2 learning: Evidence from Hebrew learners whose first language is either Semitic or Indo-European

The present study examined visual word recognition processes in Hebrew (a Semitic language) among beginning learners whose first language (L1) was either Semitic (Arabic) or Indo-European (e.g. English). To examine if learners, like native Hebrew speakers, exhibit morphological sensitivity to root and word-pattern morphemes, learners made an off-line graded lexical decision task on unfamiliar letter strings. Critically, these letter strings were manipulated to include or exclude familiar Hebrew morphemes. The results demonstrate differential morphological sensitivity as a function of participants’ language background. In particular, Indo-European-L1 learners exhibited increased sensitivity to word-pattern familiarity, with little effect of root familiarity. In contrast, Semitic-L1 learners exhibited non-additive sensitivity to both morphemes. Specifically, letter strings with a familiar root and a familiar word-pattern were the most likely to be judged as real words by this L1-Semitic group, whereas strings with a familiar root in the absence of a familiar word-pattern were the most likely to lead to a non-word decision. These findings show that both groups of learners activate their morphological knowledge in Hebrew in order to process unfamiliar Hebrew words. Critically, the findings further demonstrate transfer of L1 word recognition processes during the initial stages of second language (L2) learning.

Two hemispheres--two networks: a computational model explaining hemispheric asymmetries while reading ambiguous words

A computational model for reading that takes into account the different processing abilities of the two cerebral hemispheres is presented. This dual hemispheric reading model closely follows the original computational lines due to Kowamoto (J Mem Lang 32:474–516, 1993) but postulates a difference in architecture between the right and left hemispheres. Specifically it is assumed that orthographic, phonological and semantic units are completely connected in the left hemisphere, while there are no direct connections between phonological and orthographic units in the right hemisphere. It is claimed that this architectural difference results in hemisphere asymmetries in resolving lexical ambiguity and more broadly in the processing of written words. Simulation results bear this out. First, we show that the two networks successfully simulate the time course of lexical selection in the two cerebral hemispheres. Further, we were able to see a computational advantage of two separate networks, when information is transferred from the right hemisphere network to the left hemisphere network. Finally, beyond reproducing known empirical data, this dual hemispheric reading model makes novel and surprising predictions that were found to be consistent with new human data.

Differences and Interactions Between Cerebral Hemispheres When Processing Ambiguous Words

It is well known that the brain (especially the cortex) is structurally separable into two hemispheres. Many neuropsychological studies show that the process of ambiguity resolution requires the intact functioning of both cerebral hemispheres. Moreover, these studies suggest that while the Left Hemisphere (LH) quickly selects one alternative, the Right Hemisphere (RH) maintains alternate meanings. However, these hemispheres are connected through the corpus callosum and presumably the exchange of information is useful. In addition, many works show that the Left Hemisphere (LH) is more influenced by the phonological aspect of written words whereas lexical processing in the Right Hemisphere (RH) is more sensitive to visual form.This distinction suggests that the interconnections between the hemispheres may be used to strengthen or correct incorrect interpretations by one hemisphere. We test this hypothesis by (I) postulating that in the Left Hemisphere (LH) orthography, phonology and semantics are interconnected while (II) the Right Hemisphere (RH), phonology is not connected directly to orthography and hence its influence must be mitigated by semantical processing (III) seeing if corrections in ambiguous word processing can be aided by information in the other hemisphere. We investigate this by complementary human psychophysical experiments and by dual (one RH and one LH) computational neural network model architecturally modified from Kawamotos (1993) model to follow our hypothesis. Since the different models have different rates of convergence, we test (III) by halting processing, and using an analogue to priming to compare the rate of convergence to a corrected semantics in the LH working alone and working with information obtained from the RH at the same point in processing. In this paper we present results of the computational model and show that (I) the results obtained from the two hemispheres separately are analogous to the human experiments and (II) the use of the RH information does indeed help such corrections.

On the superiority of defaultness: Hemispheric perspectives of processing negative and affirmative sarcasm

ABSTRACT Defining defaultness in terms of an unconditional, automatic response to a stimulus allows the Defaultness Hypothesis to predict the speed superiority of default over nondefault counterparts. Here we examined the relative contribution of the cerebral hemispheres to the processing of default versus nondefault interpretations (of Hebrew items). Participants performed a lexical decision task on lateralized probes (messy) related to the default/nondefault sarcastic interpretation of their preceding negative/affirmative targets (He is/He is not the most organized student). Consistent with the Defaultness Hypothesis, probes were easier to identify in the default-negative than in the nondefault-affirmative condition. However, this superiority was more pronounced in the left hemisphere (LH) than in the right hemisphere. In particular, whereas both hemispheres reflected the superiority of default negative sarcasm over nondefault affirmative sarcasm when both targets were embedded in equally strong sarcastic contexts (Experiment 2), only the LH exhibited this very same superiority when the negatives and affirmatives were presented in isolation (Experiment 1).