Contextual Diversity IN THE Learning OF Concrete AND Abstract Words PDF

Preview

Summary

The studies on language learning in the different stages of the life cycle are innumerable, and in turn these are updated generating new proposals....

Description

1

CONTEXTUAL DIVERSITY IN THE LEARNING OF CONCRETE AND ABSTRACT WORDS The studies on language learning in the different stages of the life cycle are innumerable, and in turn these are updated generating new proposals. One of the factors with the most robust effects on the access and organization of words is the frequency with which they appear (Cuetos, 2015). The frequency effect has been consistently replicated across multiple paradigms and experimental tasks (lexical decision, semantic categorization, fixation times during eye movement tracking, etc.). The results show that the words that are repeated the most are easier to identify than those of low frequency (Broadbent, 1967; Howes & Solomon, 1951).

Contextual diversity effect Recent studies have analyzed the effect of contextual diversity, the number of different contexts in which a word appears or situations in which said lexicon is used (eg books, movies, etc.), on access to the lexicon (lexical decision times , reading times, etc.,). The results show that contextual diversity better predicts performance than frequency of use, in both adults and children. The first work to address the study of contextual diversity was carried out by Adelman, Brown, and Quesada (2006), using a naming task and a lexical decision task. The results of both tasks showed that the words that appeared in more contexts were answered more quickly. What's more, In recent studies in which the eye movement tracking technique has been used during sentence reading (eg Plummer, Perea & Rayner, 2014) similar results have also been obtained. In the study by Plummer et al. (2014) participants read sentences with experimental words that varied in terms of frequency of use and contextual diversity (data obtained from a corpus of words). The times in the

Reading were longer for words with low contextual diversity compared to those with high diversity. Furthermore, there were no differences in reading times for high and low frequency words, when controlling for contextual diversity. On the other hand, there is also empirical evidence of the effect of contextual diversity in learning new words. In a recent work, Jones, Johns, and Recchia (2012) proposed semantic distinctiveness, referring to the number of semantically distinctive contexts, rather than the mere count of documents in which a word appears, as a determining factor of learning. To measure semantic distinctiveness, they quantified the similarity of a pair of documents as the proportion of semantic information shared by both contexts (i.e., redundant information). By selecting words from a corpus, these authors found that lexical decision times were faster for those that appeared in semantically different contexts than for those that occurred in redundant contexts. To test the causal effect of contextual diversity Jones et al. (2012) also conducted an artificial language learning experiment, in which they independently manipulated the number of contexts in which an experimental stimulus appeared (i.e., the document count) and the semantic distinctiveness of these contexts. The results showed shorter latency times in a lexical decision task, for stimuli that appeared in a large number of contexts, but only when they were semantically distinctive. To explain this effect, Jones et al. (2012) proposed the Semantic Distinctivity Model according to which contexts that provide redundant information with past experiences are not coded with the same intensity as those that provide novel information. More recently, Johns, Dye, and Jones (2015) conducted an experiment in which participants read short fragments of texts extracted from magazines, books, newspapers, etc., in which pseudowords appeared instead of the corresponding very low words. frequency. These pseudowords (experimental stimuli) could appear in 5 semantically different fragments (high diversity) or in 5 semantically redundant fragments (low diversity). Subsequently, they carried out a lexical decision task and a semantic similarity judgment. The results showed a better performance for the high diversity condition in the lexical decision task, contrary to what happened in the semantic similarity trial.

The authors concluded that redundant contexts provide more stable semantic representations. Despite the importance of testing the effects of contextual diversity on the development of readers, previous research has been very scarce and restricted in terms of direct manipulation of contextual diversity with children. Since contextual diversity has been shown to be a facilitating factor for adult readers during word learning, its effect should be even greater for developing readers, considering that their word representations are less well established. In this regard, Rosa, Tapia and Perea (2017) carried out an experiment in which they experimentally manipulated contextual diversity (understood as semantic diversity) during incidental learning of new words, through the reading of texts, with children of 5th grade. primary.

Concrete effect During the first years of life, language is dominated by concrete words because experience is marked by the sensory-motor relationship with the material world, which is reflected in Piaget's theory of the four stages of cognitive development (1967 ). Specific words are characterized by having sensory aspects, their meaning refers to palpable, visible things that can be experienced through the senses, which is why they can quickly produce mental images of their content (high imaginability). Abstract words, on the other hand, do not refer to physical objects and for the most part they do not easily evoke mental images, they have low imaginability (Jefferiesm Patterson, 2009). Therefore, the meaning of abstract words is acquired through their use in different linguistic contexts, through verbal associations. The results of behavioral measurements show an advantage in the processing of concrete concepts (James, 1975; Kroll & Merves, 1986), called the concrete effect. This effect is evident in experimental tasks with shorter response times and

greater accuracy for concrete concepts (Paivio, 1991). A viable explanation for this effect could be the advantage of processing concrete concepts, having a defined sensory referent, contextual information and a greater number of semantic characteristics added to the lexical label (Paivio, 1986; Plaut & Shallice, 1993; Schwanenflugel & Shoben, 1983). To understand the development of the mechanisms underlying the learning of specific words, we can resort to the dual coding theory proposed by Clark and Paivio (1991). This theory explains human behavior and experience in terms of dynamic associative processes, operating in a rich network of modality-specific verbal and non-verbal representations. Dual coding theory postulates that cognition occurs in two independent but connected codes: a verbal code for language and a non-verbal code for mental images (Sadoski, 2005). Abstract language has less access to non-verbal images (for example, the word truth), whereas concrete language has direct sensory referents (for example, the word dog). The abstract word truth is largely defined by other language units or associations, such as non-false, objective, real, true, and exact. Abstract words are difficult to visualize and therefore less likely to be dual encoded (Clark and Paivio, 1991). Therefore, concrete language has a natural advantage over abstract language because they can be more easily represented and processed by two different codes (Sadoski, 2005). In turn, the processing of abstract nouns is based on a verbal code that would be located in the left cerebral hemisphere, while the processing of concrete nouns additionally accesses a second image-based processing system in the right hemisphere (Paivio, 1986 ). Supporting this theory, In various fields of cognitive research (Villardita, 1988), the dual coding theory attributes the advantage to the access of an image-based system in the right hemisphere. Through a study (Jessen, Heun, Erb, Granath, & Klose; 2000), using FMIR (magnetic resonance imaging works), related to events to identify the brain regions that facilitate the concrete effect, greater activation was found in the lower right and left parietal lobes, in the lower left frontal lobe, and in the precuneus, during the encoding of concrete nouns, compared to abstract nouns, which had a greater activation in the lower left frontal gyrus, (Broca's region) and in the right lateral occipital gyrus; of this dual encoding theory attributes the advantage to access to an image-based system in the right hemisphere. Through a study (Jessen, Heun, Erb, Granath, & Klose; 2000), using FMIR (magnetic resonance imaging works), related to events to identify the brain regions that facilitate the concrete effect, greater activation was found in the lower right and left parietal lobes, in the lower left frontal lobe, and in the precuneus, during the encoding of concrete nouns, compared to abstract nouns, which had a greater activation in the lower left frontal gyrus, (Broca's region) and in the right lateral occipital gyrus; of this dual encoding theory attributes the advantage to access

to an image-based system in the right hemisphere. Through a study (Jessen, Heun, Erb, Granath, & Klose; 2000), using FMIR (magnetic resonance imaging works), related to events to identify the brain regions that facilitate the concrete effect, greater activation was found in the lower right and left parietal lobes, in the lower left frontal lobe, and in the precuneus, during the encoding of concrete nouns, compared to abstract nouns, which had a greater activation in the lower left frontal gyrus, (Broca's region) and in the right lateral occipital gyrus; of this Through a study (Jessen, Heun, Erb, Granath, & Klose; 2000), using FMIR (magnetic resonance imaging works), related to events to identify the brain regions that facilitate the concrete effect, greater activation was found in the lower right and left parietal lobes, in the lower left frontal lobe, and in the precuneus, during the encoding of concrete nouns, compared to abstract nouns, which had a greater activation in the lower left frontal gyrus, (Broca's region) and in the right lateral occipital gyrus; of this Through a study (Jessen, Heun, Erb, Granath, & Klose; 2000), using FMIR (magnetic resonance imaging works), related to events to identify the brain regions that facilitate the concrete effect, greater activation was found in the lower right and left parietal lobes, in the lower left frontal lobe, and in the precuneus, during the encoding of concrete nouns, compared to abstract nouns, which had a greater activation in the lower left frontal gyrus, (Broca's region) and in the right lateral occipital gyrus; of this Greater activation was found in the lower right and left parietal lobes, in the lower left frontal lobe, and in the precuneus, during the coding of concrete nouns, compared to abstract nouns, which had greater activation in the lower left frontal gyrus. (Broca's region) and in the right lateral occipital gyrus; of this Greater activation was found in the lower right and left parietal lobes, in the lower left frontal lobe, and in the precuneus, during the coding of concrete nouns, compared to abstract nouns, which had greater activation in the lower left frontal gyrus. (Broca's region) and in the right lateral occipital gyrus; of this

In this way, it was shown that patients with lesions in the right hemisphere had a worse performance remembering specific nouns previously studied (Jessen, et al., 2000). However, these results contradict the current concept of mental image generation, which says that image activation would be expected in the occipital lobe and the inferior temporal lobe (D´Esposito, M., Detre, JA, Aguirre, GK , Stallcup, M 1997). A challenge for the dual coding model is that this model, in the study carried out by (Jessen, et al., 2000) could not explain why the concretion effect was eliminated when the stimuli were integrated into contextually rich materials, specifically, concrete words were better remembered than abstract words when they were presented in sentences in random order and in isolation; However, when the sentences are presented in a coherent paragraph, there were no differences in the recall of the concrete and abstract words. (Jessen, et al., 2000) conclude that the concreteness effect cannot be explained exclusively by the dual coding theory, but through the combination of the two models that explain this effect, dual coding and context availability. As an alternative to dual encoding theory, context availability theory argues that faster recognition of concrete versus abstract words is due to greater contextual support and not to a non-verbal encoding system (Kieras, 1978 ; Schwanenflufel, Akin & Luch 1992; Schwanenflugel et al., 1988; Schwanenflugel & Shoben 1983; Van Hell & de Groot, 1998, 2008). Since abstract words are often found in a wide variety of contexts, their contextual information is represented more flexibly, and their poor performance is due to the relative unavailability of such associated information in memory. Schwanenflugel and Stowe (1989) propose an explanation in line with this theory. Concrete names trigger more associative information, resulting in faster recognition of these items. However, if abstract names are presented in a meaningful context with sufficient verbal information, the concretion effect disappears and abstract names are recognized as quickly as concrete names. The context availability model was discussed by Kieras (1978), the model establishes that comprehension processes in language are supported by adding contextual information to the materials that must be understood, such contextual information can come from the stimulus environment or knowledge of the subject, allowing to configure the necessary relationships between the concretion effect disappears and abstract names are recognized as quickly as concrete names. The context availability model was discussed by Kieras (1978), the model establishes that comprehension processes in language are supported by adding contextual information to the materials that must be understood, such contextual information can come from the stimulus environment or knowledge of the subject, allowing to configure the necessary relationships

between the concretion effect disappears and abstract names are recognized as quickly as concrete names. The context availability model was discussed by Kieras (1978), the model establishes that comprehension processes in language are supported by adding contextual information to the materials that must be understood, such contextual information can come from the stimulus environment or knowledge of the subject, allowing to configure the necessary relationships between

the concepts of the new message, therefore understanding is deeply related to the person's ability to provide context for the linguistic message. (Bransford & McCarrell, 1974). Abstract words presented in isolation are not as easily understood as concrete words, because people have more difficulty determining the appropriate contextual information for abstract materials (Holmes and Langfard 1976). In learning, this understanding deficit leads to less detailed representation in memory of abstract materials. In the research of Schwanenflugel and Shoben (1983) the objective was to analyze the incidence of the context at the time of understanding abstract and concrete words, and develop a theoretical perspective with which the effects of the context prior to storage in the memory. They carried out several experiments, in one experiment they presented abstract and concrete sentences, with and without paragraph context. (For example, A Sentence with Context: A company was rapidly beginning to fail, managers realized that something had to be done soon. After several weeks they would develop a strategy to attract customers. Abstract statement: The goal was to double profits Verification of the judgment: The company was failing) Without a context, the subjects took longer to read the abstract sentences than the concrete sentences. With context, the reading times between concrete and abstract did not differ. In the second experiment, similar results were obtained, in a lexical decision task, reaction times were measured in the absence of context, the lexical decision was faster for concrete words than for abstract words. However, in the context of a sentence the lexical decision times for both conditions were equivalent. The results were discussed in support of the context availability model. (Schwanenflugel et al., 1983) In a lexical decision task, reaction times were measured in the absence of context, the lexical decision was faster for concrete words than for abstract words. However, in the context of a sentence the lexical decision times for both conditions were equivalent. The results were discussed in support of the context availability model. (Schwanenflugel et al., 1983) In a lexical decision task, reaction times were measured in the absence of context, the lexical decision was faster for concrete words than for abstract words. However, in the context of a sentence the lexical decision times for both conditions were equivalent. The results were discussed in support of the context availability model. (Schwanenflugel et al., 1983) In another study (Taylor, Francis, Vazquez, & Carbajal, 2018), they measured whether the availability of context contributed to the effects of concretion on explicit memory. It is known that the performance of verbal memory varies according to the semantic and lexical characteristics of the word to be remembered; the concretion of words has marked effects on

explicit memory tasks (eg, Balota & Neely, 1980; Glanzer & Adams, 1985; Holmes & Langford 1976). The benefits of context can come from accessing previous experiences or from interacting with a new context (Schwanenflugel et al.

al., 1988). One explanation for why concrete words are remembered more than abstract words is the systematic differences between these types of words in the availability of context information. In the study by Teylor et al. (2018) investigated the extent to which the availability of context can explain the effects of word concretion. In two experiments, they measured a free recall task (exp 1) and a recognition task (exp 2), randomly presenting object words in a context of low semantic restriction and high semantic restriction; the semantic restriction (understood as the amount of information that is provided in a sentence that helped to infer the meaning of the object word), and the effects of this with verbal memory; specificity and frequency of words, In each experiment the study task was to read words or phrases aloud and try to memorize the target words. There were significant results of concreteness and context in both memory tasks, but the magnitude of the results of the concreteness effect did not differ for words studied in contexts of sentences with high or low restriction. The results only support the hypothesis of context significance for lexical processing in explicit memory performance (Teylor et al., 2018) Previous studies provide information regarding the process of metal representation of concrete words, but little is known about the internalization of abstract words. As stated by Jones and Recchia (2012), abstract words have a great representation within the words that we know and many of them are used very frequently, but there is little empirical work on this topic, resulting today in a challenge for the investigation. In the work of Zdrazilova and Pexman (2013) they extended the semantic richness approach to abstract words, analyzing the effects of six variables on lexical-semantic processing for 207 abstract nouns. They used different semantic dimensions: availability of context, emotional sensory-motor experience, arousal, semantic neighborhood, number of associations (associative network). Regarding sensory experience, although abstract words do not refer to concrete objects, Dicke, Taylor, and Gullick (2011) argue that it is possible to apprehend their mean...