Brain Advance Access originally published online on August 5, 2003
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Brain, Vol. 126, No. 11, 2419-2430,
November 2003
© 2003 Guarantors of Brain
doi: 10.1093/brain/awg243
Idiom comprehension in Alzheimer’s disease: the role of the central executive
1 Dipartimento di Psicologia, Università di Milano-Bicocca, 2 Divisione di Neurologia, Azienda Ospedaliera Ospedale Niguarda Ca’ Granda, Milano, 3 Dipartimento di Psicologia, Università di Palermo, Palermo, Italy
Correspondence to: Costanza Papagno, Dipartimento di Psicologia, Università di Milano-Bicocca, Piazza dell’Ateneo Nuovo 1, Edificio U6, 20126 Milano, Italy E-mail: costanza.papagno{at}unimib.it
Received January 30, 2003. Revised March 26, 2003. Second revision May 22, 2003. Accepted May 26, 2003.
 |
Summary |
|---|
 Top Summary IntroductionExperiment 1Experiment 2General discussionAppendixReferences |
|---|
Idiom comprehension of 15 patients with mild probable Alzheimer’s disease was examined by means of a sentence-to-picture matching task. Patients had to choose between two pictures, one representing the figurative and the other the literal interpretation. They were also submitted to a literal sentence comprehension test and to a pencil-and-paper dual task. Whereas literal comprehension was normal in seven subjects and mildly impaired in the others, idiom comprehension was very poor in all of them and correlated with the performance on the dual task. When the idiom test was repeated using an unrelated situation as an alternative to the picture representing the figurative meaning, performance significantly improved. It was hypothesized that the response in the sentence-to-picture matching task in the case of idioms requires sentence processing followed by the suppression of the literal interpretation. Alzheimer’s disease patients proved to be unable to inhibit the literal meaning, although they had not lost the idiomatic meaning. In a second experiment, 15 Alzheimer’s disease patients with a comparable level of cognitive impairment were submitted to the same idiom comprehension test, and to a test of verbal explanation of the idioms. The results showed significantly better performance in the oral task than in the sentence-to-picture matching task. In oral explanation, however, Alzheimer’s disease patients also produced some literal interpretation whenever this represented a possible situation in the real world. We suggest that, during idiom interpretation, the literal meaning needs to be suppressed in order to activate the figurative meaning, and we stress the fact that both linguistic and extralinguistic factors must be taken into account to explain idiom interpretation.
Keywords: idiom; central executive; suppression; Alzheimer’s disease
Abbreviations: MODA = Milan Overall Dementia Assessment
|
Introduction |
|---|
|
TopSummaryIntroductionExperiment 1Experiment 2General discussionAppendixReferences |
|---|
An aspect of linguistic communication is the use of figurative expressions, the meaning of which differs from the literal meaning of their component words; examples of these expressions are proverbs, hyperbole, metaphors and idioms. Two aspects concerning figurative language have been studied in detail: linguists have addressed problems related to the mechanisms involved in the interpretation of figurative language, and neuropsychologists have been concerned with its neural correlates. These expressions, however, do not constitute a homogeneous group, and idioms, metaphors, proverbs etc. have different characteristics. Idioms are among the most common forms of figurative language. They are typically described as frozen phrases whose meanings are stipulated directly in a mental lexicon, and the speaker’s meaning cannot be derived from an analysis of the words’ typical meanings. Many idioms, however, are more complex and dynamic (Glucksberg et al., 1993
). They range from expressions that are almost like long words (e.g. ‘by and large’) to expressions that are like metaphors (e.g. ‘skate on thin ice’). In between there is the majority of idioms, in particular verbal idioms, that are processed syntactically and semantically, and accordingly can undergo syntactic and semantic variations.
Neuropsychological studies on figurative language comprehension in brain-damaged patients concern mainly metaphors. There is, however, some information available on idioms, suggesting that patients with left hemisphere damage outperform right-brain-damaged patients in figurative sentence comprehension (Foldi et al., 1983; Van Lancker and Kempler, 1987; Kempler and Van Lancker, 1993; Burgess and Chiarello, 1996), even though psycholinguistic models all stress the importance of lexical knowledge (Swinney and Cutler, 1979; Tabossi and Zardon, 1995) and increasing evidence indicates the involvement of syntactic processing (Peterson et al., 2001) in the comprehension of idioms.
There are, in our opinion, many limitations in previous neuropsychological studies on idioms. First, even if the literal and non-literal sentences were matched in structure, items were not chosen on the basis of their intrinsic linguistic features, and idioms, proverbs and courtesy phrases were considered together as ‘familiar language’ (Van Lancker and Kempler, 1987; Kempler et al., 1999). The severity of aphasia and the modalities of assessment are not always described; the time of testing with respect to the time of onset of aphasia was variable and it is not clear whether patients underwent language rehabilitation; other neuropsychological deficits are not reported; and the number of stimuli was limited. Moreover, in one of these studies (Van Lancker and Kempler, 1987), the performance of aphasic patients cannot be considered normal since their score was 
72% correct for idioms, compared with the controls’ performance of 97.3%. The same aphasic subjects scored 90% correct on word comprehension.
Papagno and colleagues (Papagno et al., 2002) and Papagno and Tabossi (2002) produced data which do not support the right hemisphere hypothesis. In both studies, a sentence-to-picture matching task was used for idiomatic and literal sentences. A homogeneous class of expressions was selected. These were non-ambiguous, opaque idioms; the meaning could not be directly derived from the figure involved and the literal interpretation was implausible because it had no meaning (for example, the idiom ‘venire alle mani’ literally means ‘to come to the hands’) (Nunberg et al., 1994). In the first study, repetitive transcranial magnetic stimulation was used to evaluate the neural correlates of idiom comprehension. It was found that only left hemisphere stimulation increased reaction times and reduced accuracy in a task in which participants had to choose between a literal and a figurative representation of the idiomatic sentence. In the second study (Papagno and Tabossi, 2002), ten aphasic left-brain-damaged subjects were submitted to the same tests. They showed a performance in idiom comprehension that was significantly impaired compared with that of a matched group of normal controls, even when aphasic deficits were mild. These difficulties seem to be due, to a large extent, to the fact that patients, unlike unimpaired listeners, rely on a literal-first strategy in the interpretation of idioms, and access a figurative interpretation only when the linguistic analysis fails to yield acceptable results (Bobrow and Bell, 1973).
As far as patients with Alzheimer’s disease are concerned, it has been suggested that their behaviour is similar to that of right-brain-damaged patients (Winner and Gardner, 1977). It is not clear which expressions were used, how severe their dementia was and how it was tested. Kempler and colleagues examined 29 patients with probable Alzheimer’s disease ranging from mild (Mini-Mental State Examination score 28) to severe (Mini-Mental State Examination score 2); they were given tests of words, idioms, proverbs (overall, ten stimuli) and novel phrase comprehension (Kempler et al., 1988). In the figurative language comprehension task, no alternative corresponding to the literal interpretation was available, but only a referential representation of one word in the stimulus (concrete response). The results showed that Alzheimer’s disease patients had difficulty interpreting abstract meanings: when faced with alternative interpretations of familiar phrases, they chose concrete responses, suggesting that they were using lexical (single-word), referential meaning to interpret the phrases (Kempler et al., 1988). Familiar phrases are considered as psychological ‘chunks’ similar to words, perceived and understood as wholes, without analysis of constituent parts taking place at all.
Although we disagree with this interpretation and suggest that idioms undergo processes very similar to those involved in the comprehension of literal expressions, in particular a full syntactic analysis (C.Papagno, P.Tabossi, M.R.Colombo, P.Zannetti, unpublished results; Peterson et al., 2001), we accept that idiom comprehension can be impaired for other reasons, apart from lexical–semantic and syntactic deficits. In a recent study on Alzheimer’s disease patients, it was shown that the decline of figurative language (assessed by means of an oral explanation task of metaphors and idioms) is not an early symptom of dementia, can be preserved longer compared with propositional language, and can occur independently of the impairment of propositional language (Papagno, 2001). Idioms were both ambiguous and non-ambiguous, decomposable and non-decomposable, frozen and unfrozen. Performance on the idiom comprehension test was contrasted with performance on the Token Test. There were, however, very few patients with an impairment of figurative language but normal propositional language. It was suggested that an idiom needs to be present as a whole in the subject’s semantic store, and then it must be correctly accessed and retrieved. Semantic memory and executive functions, therefore, seem to be involved. Consistent results were found by Papagno and Vallar (2001), investigating the ability of a patient with Down syndrome to appreciate non-literal and literal aspects of language. Whereas the latter was largely preserved, the former was defective. Some executive aspects of working memory were also impaired. What could be taken for certain in this patient was that impaired appreciation of metaphor and idioms could not be interpreted in terms of syntactic and lexical–semantic aspects of language comprehension, but might involve executive functions.
Patients with Alzheimer’s disease have impaired executive functions, even in the early stage (Baddeley et al., 1997). Therefore, we decided to explore idiom comprehension in this group of subjects, given that Alzheimer’s disease patients with no deficit in propositional language comprehension but impaired figurative language comprehension have been found (Papagno, 2001). In these cases (impaired idiom comprehension, normal literal comprehension), difficulties in idiom comprehension cannot be attributed to lexical–semantic and syntactic deficits.
In particular, we examined the relation between idiomatic and literal language in Alzheimer’s disease patients and the role of executive functions in idiom comprehension.
|
Experiment 1 |
|---|
|
TopSummaryIntroductionExperiment 1Experiment 2General discussionAppendixReferences |
|---|
Subjects
Fifteen patients (12 women and three men, mean age 71.73, SD 8.86; years of education 8.6, SD 4.12) with probable Alzheimer’s disease were selected on the basis of their clinical history, their CT or MRI findings and their performance on the Milan Overall Dementia Assessment (MODA; Brazzelli et al., 1994
). Only patients with a score >63/100, which corresponds to a mild degree of cognitive impairment, were included in the study. Other criteria of inclusion were (i) the absence of previous neurological diseases, (ii) the absence of lesion on the CT scan, and (iii) a minimum of 3 years of education. Demographic and clinical data for the 15 Alzheimer’s disease patients are reported in Table 1. They all gave informed consent to participate in the study.
|
Procedure
Forty unambiguous verbal idioms were selected (see Appendix). They were rated as familiar by a group of 45 healthy subjects from different parts of Italy (31 women and 14 men, age 22–65 years), who had to judge whether each idiom was unknown (score 0), known but not sure about the meaning (score 1), familiar, with a well-known meaning (score 2), or highly familiar (score 3).
Idioms were unambiguous: either they were ill-formed from a syntactical point of view (some Italian idioms contain a syntactic violation) or their literal interpretation was unlikely, because it made no sense (as in the example above, ‘venire alle mani’) or the situation was implausible in the real world (e.g. ‘spremersi le meningi’, which means ‘to squeeze one’s own meninges’). In a first version, for each idiom, two line drawings were created, one representing the idiomatic interpretation and the other representing as well as possible the literal interpretation (Fig. 1A). In a second version, the test was administered again with two alternatives, but the picture representing the literal interpretation was replaced by one representing an unrelated situation, which was defined by a sentence containing one word of the idiomatic expression (for example, when the idiomatic expression was ‘to have a green thumb’, the alternative was a man hurting his thumb) (Fig. 1B). Each patient performed both versions.
|
Tests
Patients performed three tasks. The order of presentation was counterbalanced across participants.
Idiom comprehension
This was a sentence-to-picture matching task: the examiner read a sentence corresponding to an idiom to the patient, who chose between two pictures. Patients’ performance on this task was compared with the performance of a group of 15 elderly healthy participants, matched for age and educational level.
Literal sentence comprehension (Parisi and Pizzamiglio, 1970)
This test included 80 items. The patient was shown a pair of pictures while one sentence was read to him. One of the pictures was correctly described by the sentence; the other corresponded to a sentence that was identical except for a detail (i.e. ‘the boy draws’ versus ‘the boy will draw’). The sentences were matched, as far as possible, in length and complexity to idiomatic sentences. As no control subject scored <70 on this task in the original study and in a subsequent one (Vallar et al., 1988), the score of 70 was taken as the cut-off point (Fig. 2).
|
Pencil-and-paper dual task (Baddeley et al., 1997)
After being tested for a baseline span (the maximum length at which the patient recalled three lists without errors), patients were presented with lists of digits at their own span for a period of 2 min. Patients were then required to cross out, in a specific sequence, boxes (1 cm square) which had been linked to form a path laid out on an A4-size sheet of white paper over a 2-min interval. Finally, the two tasks were combined (digit span and tracking) for 2 min. A score combining the performance on the two tasks was given (control mean = 91.99 ± 11.10). This test was chosen because it proved to be particularly sensitive to early dysexecutive changes in Alzheimer’s disease patients.
Results
The results for the 15 patients in the three tests are shown in Table 2. The mean number of correct responses in idiom comprehension when a literal foil was used was 20.67 (SD 7.97), whereas for the 15 healthy participants the mean number of correct responses was 34.53 (4.37). The t-test showed that the difference between Alzheimer’s disease and control subjects was highly significant [t(14) = –5.963, P < 0.0001]. Sentence comprehension proved to be preserved in seven patients and mildly impaired in the others, with the exception of Patients 6 and 8, whose impairment was moderate. Performance on idiom comprehension was correlated with that on the dual-task (rS = 0.89, P < 0.001), whereas there was no correlation between the former and the degree of cognitive decline, as assessed by the MODA (rS = –0.26, P = 0.92) or sentence comprehension (rS = 0.24, P = 0.37). Finally, sentence comprehension was not correlated either with dual-task performance (rS = 0.16, P = 0.54) or with cognitive decline (rS = 0.41, P = 0.13). We submitted 20 normal subjects to a rating task to check the plausibility of the pictures representing the literal interpretations. The subjects were asked to rate how well the picture corresponding to the literal interpretation of the idiom matched the sentence read by the examiner. A score ranging from 0 to 7 was assigned, where 0 meant ‘not at all’ and 7 ‘perfectly’. A correlation analysis indicated that the more implausible an idiomatic sentence is as a linguistic description of the literal picture, the more likely it is that, upon presentation of that sentence, the patient will select the idiomatic picture (rS = –0.32, P < 0.05). In contrast, there was no effect of familiarity in this task (rS = 0.02, P = 0.87).
|
When the test was re-administered in the second version (one picture representing the idiomatic interpretation and the other representing an unrelated situation including a word contained in the idiom), performance significantly improved (mean of correct responses 33.13, SD 5.20) [t(14) = –6.38, P < 0.0001], showing that patients had a good knowledge of the meaning of the idiom, despite their poor performance in the previous task. Every item that was selected correctly when the alternative was the literal interpretation was also chosen correctly when the alternative was the unrelated situation, demonstrating a consistent response of the patient, who was not performing at chance.
Discussion
Patients chose the literal interpretation when the corresponding picture was highly plausible but chose the idiomatic interpretation when the alternative represented an unrelated situation, even when the distractor corresponded to a sentence including a word contained in the idiom. This means that Alzheimer’s disease patients know the meaning of the idiom, but the literal interpretation, which is also activated, strongly interferes with it; it also means that patients are not picking the literal interpretation on the basis of a single word in the idiom. It seems as if patients are unable to suppress the literal interpretation when there is an overt representation of it. We suggest that the activation of the literal interpretation is stronger or perhaps quicker than the activation of the idiomatic meaning, as hypothesized by Bobrow and Bell (1973), in contrast with the conclusion drawn with normal subjects (Gibbs, 1980). The interference produced by the literal interpretation is similar to that observed in the Stroop test: the choice of the literal representation can be due to the greater automaticity of the inhibition of the response with respect to that actually requested (MacLeod, 1991). It has been demonstrated that Alzheimer’s disease patients produce significantly larger interference effects with the Stroop Color-Word Test than comparable normal people (Bondi et al., 2002)
To further investigate this possibility (interference of the literal interpretation), we selected another group of Alzheimer’s disease patients, comparable with respect to their degree of cognitive impairment, and we submitted them to a second experiment. The hypothesis was that when a verbal explanation is required, people are not ‘offered’ an external literal interpretation but generate their own response internally. Indeed, the representation of the literal meaning could reinforce the literal interpretation, which seems to take place. If the idiomatic meaning were unknown, a greater number of unrelated responses would be expected, the literal ones being produced only in the case of a highly plausible meaning.
|
Experiment 2 |
|---|
|
TopSummaryIntroductionExperiment 1Experiment 2General discussionAppendixReferences |
|---|
Subjects
Fifteen Alzheimer’s disease patients, five men and 10 women (mean age 75.53 years, SD 7.98, range 56–85; years of education 7.93, SD 4.26, range 3–17), entered the study. They were comparable to the patients of the previous group in severity of disease [t(14)= –1.353, P = 0.19], mean age [t(14)= –1.225, P = 0.24] and education [t(14) = 0.406, P = 0.69]. The same criteria of inclusion were followed. Demographic and clinical data of the patients are reported in Table 3. All patients gave informed consent to participation in the study.
|
Procedure
In this experiment patients were subjected to the dual task, to the idiom comprehension test in the sentence-to-picture matching modality, and to a test of oral explanation of idioms. In this last task, the same idioms were presented verbally to the patient, who was asked to explain their meanings (after being instructed that he/she was expected to give a figurative interpretation of the sentence). The oral explanation was accepted when at least two out of three independent judges considered it adequate. The order of tasks was counterbalanced across subjects. We assumed that when Alzheimer’s disease patients have simply to explain the meaning of an idiom, they attain a better performance than when they have to operate a choice between two alternatives, where the concrete representation interferes with the retrieval of the abstract interpretation. In other words, the lexical knowledge of idioms is preserved, but patients are not able to reject the literal meaning when this is offered to them.
Results
The results for the 15 patients in the three tests are shown in Table 4. Their performance was significantly worse than that of 15 healthy participants, matched for age and educational level, both in the sentence-to-picture matching task [t(14) = –5.377, P < 0.0001] and in the oral task [t(14) = –6.994, P < 0.0001]. However, the performance of the Alzheimer’s disease patients was significantly better in the oral explanation task (mean number of correct responses 26.27, SD 5.57) compared with their performance in the sentence-to-picture matching task (mean number of correct responses 19.4, SD 8.84) [t(14) = –3.17, P < 0. 01]. Whereas for 35.5% of the total answers the patients gave a correct response in both tasks, 31.2% of answers were correct in the oral task and wrong in the sentence-to-picture matching task. Responses were wrong in both tasks in 20.5%, and only 12.8% of responses were correct in the sentence-to-picture and wrong in the oral task.
|
As in the previous experiment, the performance on idiom comprehension in the sentence-to-picture matching task correlated with that on the dual task (rS = 0.54, P < 0.05), while there was no correlation between the former and the degree of cognitive decline (rS = –0.29, P = 0.28). Also, performance in the oral task did not correlate with cognitive decline (rS = 0.22, P = 0.41) or dual-task performance (rS = 0.27, P = 0.30). The number of correct responses correlated significantly with literal picture plausibility in the sentence-to-picture matching task (rS = –0.35, P = 0.02). The number of correct responses in the oral task correlated significantly with familiarity (rS = 0. 42, P < 0.01).
We also checked sentence plausibility, because some oral responses corresponded to the literal interpretation (25%). Ten normal controls were asked to rate how well an idiomatic string would also evoke a literal interpretation. For example: the sentence ‘to go to the cool’ means in Italian ‘to be put in jail’, but one could accept the possibility that someone is moving towards a cool place, while the sentence ‘to squeeze his/her own meninges’ does not allow any literal explanation. Controls had to give a rating of 0 (no acceptable literal meaning), 1 (mild literal meaning) or 2 (absolutely acceptable literal meaning). A correlation analysis showed that Alzheimer’s disease patients produced the literal interpretation whenever the sentence offered this possibility: this means that, whenever plausible, the literal meaning overpasses the figurative one (rS = 0.63, P < 0.0001). The remaining errors in the oral task were mainly partial or wrong interpretation (57%), no response (17%) and opposite meaning (1%).
Discussion
The performance of Alzheimer’s disease patients was significantly better in the oral task than in the sentence-to-picture matching task. In this last task performance correlated with the plausibility of the pictures, whereas in the oral task performance correlated with familiarity. A correlation was also found between the literal errors produced in the oral task and the degree of plausibility of the sentence, suggesting that, at least out of context, the literal interpretation precedes the figurative one for patients with Alzheimer’s disease, inhibiting further analysis whenever the result of sentence processing is plausible.
|
General discussion |
|---|
|
TopSummaryIntroductionExperiment 1Experiment 2General discussionAppendixReferences |
|---|
Two experiments were conducted with Alzheimer’s disease patients. In Experiment 1, an analysis of the relationship between idiomatic and literal language was performed. Idiomatic comprehension was tested in two conditions, both using a sentence-to-picture matching task paradigm: a picture representing the idiomatic meaning was contrasted, in one condition, with a picture representing the literal meaning, whereas in the other condition it was contrasted with a picture representing an unrelated situation, including a word contained in the idiom. There was no correlation between idiomatic and literal language comprehension, as already found in a previous study (Papagno, 2001
). However, participants had better idiomatic comprehension than is reflected in the first condition (idiomatic versus literal interpretation): indeed, their performance significantly improved with the unrelated alternative. In Experiment 2, idiom comprehension was assessed in two modalities: patients performed a sentence-to-picture matching task (two alternatives: correct figurative meaning versus literal representation) and an oral task in which they produced a verbal explanation of the idioms. This latter task was significantly better performed than the former, but patients still produced a literal interpretation whenever the literal meaning of the sentence was plausible.
In both experiments, performance on idiomatic expressions correlated with performance on the dual task.
The conclusions of this study are straightforward. First, our data confirm those of a previous study (Papagno, 2001), in which figurative language comprehension was dissociated from literal language comprehension, but also suggest that the modality of testing is very important, as already outlined by Tompkins and colleagues (Tompkins et al., 1992). In Papagno’s (2001) paper, idioms were both ambiguous and non-ambiguous and varied in transparency, leading to more literal interpretations in the oral modality. In the present study, idioms were chosen in order to be non-ambiguous, and the literal interpretation in the oral task is therefore restricted to the few plausible ones. Kempler and colleagues used ten stimuli, which they called ‘familiar phrases’, generically referring to overlearned expressions, including proverbs, idioms and social interaction formulae, some of the items being full sentences and others being phrases (Kempler et al., 1988). We have already discussed the fact that even restricted categories of idioms can be divided into subgroups with different features, and therefore it is dangerous to consider them together. There is another point to discuss: Kempler and colleagues stated that the oral type of testing often presents problems because responses can be judged to be poor, as Alzheimer’s disease patients may choose the wrong word or may not respond to the question asked, or the response may be uninterpretable (Kempler et al., 1988). For this reason these authors required a picture pointing response, suggesting that deficient performance can be interpreted more clearly with this type of response. Our patients, however, had only mild cognitive impairment and, more importantly, they had no remarkable language deficits. Moreover, we accepted all responses, that showed that the core abstract meaning of the idiom was preserved. On the other hand, we have found that a sentence-to-picture matching task can produce other problems.
Another point deserves further discussion. In the previous study by Papagno (2001), out of 39 Alzheimer’s disease patients, only two were found with an impairment of metaphor and idiom comprehension and preserved literal comprehension, while six showed the opposite pattern. This could be due to the fact that poor literal comprehension is not sufficient or necessary to reduce figurative interpretation. Idiomatic language can be impaired not only because of lexical–semantic and syntactic deficits, as observed in aphasic patients (Papagno and Tabossi, 2002), but can be produced by different mechanisms. One possibility is inadequate suppression, as seen in the Stroop task. Brega and Healy (1999) showed that sentence processing can be obligatory when the component words are highly relevant to the task. In their experiment, participants showed longer reaction times and more errors in naming the colours in which words were printed when the words were included in a sentence and were colour words or colour-related words. By analogy, one can assume that idioms are processed syntactically and semantically and that this produces the interfering effect. Gernsbacher and Robertson (1999) had already described the crucial role that suppression plays in many aspects of language comprehension. They define suppression as a general, cognitive mechanism, the purpose of which is to attenuate the interference caused by the activation of extraneous, unnecessary or inappropriate information. Sometimes this superfluous activation arises from the external environment, as when we conduct a conversation in a noisy place. At other times this superfluous information is activated internally, as when we have to deal with the competing meanings of a word or phrase. In the case of idiomatic expressions, the literal and figurative interpretations can be activated in parallel, so that a mechanism for attenuating the activation of the inappropriate interpretation is needed. However, it has been proposed that there are some asymmetries in what gets activated and what gets suppressed. Giora and Fein (1999) suggested that hearing familiar idioms should lead to both idiomatic and literal interpretations becoming activated, because both interpretations are salient out of context. In contrast, less familiar idioms are more likely to activate a literal interpretation, out of context. According to their view, the literal interpretation of an idiom is functional for idiom interpretation. Therefore they predict that, with familiar idioms, deriving the literal interpretation does indeed involve suppressing the idiomatic interpretation; however, they propose that deriving the idiomatic interpretation requires the literal interpretation to be retained. Our data do not support this hypothesis, given that familiar idioms need suppression of the literal interpretation in order to derive the idiomatic meaning. It seems as if familiar idioms, out of context, are also more likely to activate the literal meaning, so that suppression is necessary.
We suggest that suppression is mediated by the central executive. In the case of an absolutely implausible expression or an expression that contains a syntactic violation (Papagno and Tabossi, 2002), the literal meaning is less salient and can be easily rejected even when the central executive is damaged, but when the literal interpretation is acceptable or is overtly suggested, suppression is necessary. As the central executive is damaged in Alzheimer’s disease patients, as already observed in previous studies (Baddeley et al., 1991, 1997), inhibition of the literal meaning does not take place efficiently and the figurative meaning does not get sufficient activation.
Finally, as a consequence of these observations, we stress that several factors have to be taken into account in studying idiom comprehension: their intrinsic features, such as their compositionality, transparency (Nunberg et al., 1994) and familiarity; linguistic aspects, such as semantic and syntactic processing; and extralinguistic abilities, such as executive and visuoperceptual functions.
|
Acknowledgements |
|---|
The study was funded through a grant of the MIUR (Ministero Istruzione, Universita Ricerca) and FAR (Finanziamento di Ateneo per la Ricerca) (60%) to C.P.
|
Appendix |
|---|
|
TopSummaryIntroductionExperiment 1Experiment 2General discussionAppendixReferences |
|---|
Where possible the equivalent English idiom is given, otherwise an explanation of the figurative meaning is given.
|
|
|
References |
|---|
|
TopSummaryIntroductionExperiment 1Experiment 2General discussionAppendixReferences |
|---|
Baddeley AD, Bressi S, Della Sala S, Logie RH, Spinnler H. The decline of working memory in Alzheimer’s disease: a longitudinal study. Brain 1991; 114: 2521–42.
Baddeley AD, Della Sala S, Gray C, Papagno C, Spinnler H. Testing central executive function with a pencil-and-paper test. In: Rabbitt P, editor. Methodology of frontal and executive function. Hove (UK): Psychology Press; 1997. p. 61–80.
Bobrow S, Bell S. On catching on to idiomatic expressions. Mem Cognit 1973; 19: 295–308.
Bondi MW, Serody AB, Chan AS, Eberson-Shumate SC, Delis DC, Hansen LA, et al. Cognitive and neuropathologic correlates of Stroop Color-Word Test performance in Alzheimer’s disease. Neuropsychology 2002; 16: 335–43[CrossRef][Web of Science][Medline]
Brazzelli M, Capitani E, Della Sala S, Spinnler H, Zuffi M. A neuropsychological instrument adding to the description of patients with suspected cortical dementia: the Milan Overall Dementia Assessment. J Neurol Neurosurg Psychiatry 1994; 57: 1510–7.
Brega AG, Healy AF. Sentence interference in the Stroop task. Mem Cognit 1999; 27: 768–78.[Web of Science][Medline]
Burgess C, Chiarello C. Neurocognitive mechanisms underlying metaphor comprehension and other figurative language. Metaphor Symbolic Activity 1996; 11: 67–84.
Foldi NS, Cicone M, Gardner H. Pragmatic aspects of communication in brain damaged patients. In: Segalowitz SJ, editor. Language functions and brain organization. New York: Academic Press; 1983. p. 51–86.
Gernsbacher MA, Robertson RRW. The role of suppression in figurative language comprehension. J Pragm 1999; 31: 1619–30.[CrossRef]
Gibbs RW. Spilling the beans on understanding and memory for idioms in conversation. Mem Cognit 1980; 8: 449–56.
Giora R, Fein O. On understanding familiar and less-familiar figurative language. J Pragm 1999; 31: 1601–18.[CrossRef]
Glucksberg S, Brown M, McGlone MS. Conceptual metaphors are not automatically accessed during idiom comprehension. Mem Cognit 1993; 21: 711–9.[Web of Science][Medline]
Kempler D, Van Lancker D. Acquisition and loss of familiar language: idiom and proverb comprehension. In: Eckman FR, editor. Confluence: linguistics, L2 acquisition, and speech pathology. Language acquisition and language disorders, Vol. 4. Amsterdam: John Benjamins; 1993. p. 249–57.
Kempler D, Van Lancker D, Read S. Proverb and idiom comprehension in Alzheimer disease. Alzheimer Dis Assoc Dis 1988; 2: 38–49.[Medline]
Kempler D, Van Lancker D, V. M, Bates E. Idiom comprehension in children and adults with unilateral brain damage. Dev Neuropsychol 1999; 15: 327–49.[Web of Science]
MacLeod CM. Half a century of research on the Stroop effect: an integrative review. Psychol Bull 1991; 109: 163–203.[CrossRef][Web of Science][Medline]
Nunberg G, Sag IA, Wasow T. Idioms. Language 1994; 70: 491–538.[CrossRef][Web of Science]
Papagno C. Comprehension of metaphors and idioms in patients with Alzheimer’s disease. A longitudinal study. Brain 2001; 124: 1450–60.
Papagno C, Tabossi P. idiom comprehension in aphasic patients. Brain Lang 2002; 83: 78–81.
Papagno C, Vallar G. Understanding metaphors and idioms: a single-case neuropsychological study in a person with Down syndrome. J Int Neuropsychol Soc 2001; 7: 516–27.[CrossRef][Web of Science][Medline]
Papagno C, Oliveri M, Romero L. Neural correlates of idiom comprehension. Cortex 2002; 38: 895–8.
Parisi D, Pizzamiglio L. Syntactic comprehension in aphasia. Cortex 1970; 6: 204–215.[Medline]
Peterson RR, Burgess C, Dell GS, Eberhard K. Dissociation between syntactic and semantic processing during idiom comprehension. J Exp Psychol Learn Mem Cogn 2001; 27: 1223–37.[CrossRef][Web of Science][Medline]
Swinney DA, Cutler A. The access and processing of idiomatic expression. J Verb Learn Verb Behav 1979; 18: 523–34.[CrossRef]
Tabossi P, Zardon F. The activation of idiomatic meaning. In: Everaert M, van der Linden EJ, Schenk A, Schreuder R, editors. Idioms: structural and psychological perspectives. Hillsdale (NJ): Lawrence Erlbaum; 1995. p. 273–82.
Tompkins CA, Boada R, McGarry K. The access and processing of familiar idioms by brain-damaged and normally aging adults. J Speech Hear Res 1992; 35: 626–37.[Medline]
Vallar G, Papagno C, Cappa S. Latent dysphasia after left hemisphere lesions: a lexical–semantic and verbal memory deficit. Aphasiology 1988; 2: 463–78.
Van Lancker D, Kempler D. Comprehension of familiar phrases by left but not by right hemisphere damaged patients. Brain Lang 1987; 32: 265–77.[CrossRef][Web of Science][Medline]
Winner E, Gardner H. The comprehension of metaphor in brain-damaged patients. Brain 1977; 100: 717–29.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||