Experimental pragmatics: a Gricean turn in the study of language

Discerning the meaning of an utterance requires not only mastering grammar and knowing the meanings of words but also understanding the communicative (i.e., pragmatic) features of language. Although it has been an ever present aspect of linguistic analyses and discussions, it is only over the last ten years or so that cognitive scientists have been investigating--in a concerted fashion--the pragmatic features of language experimentally. We begin by highlighting Paul Grice's contributions to ordinary language philosophy and show how it has led to this active area of experimental investigation. We then focus on two exemplary phenomena--'scalar inference' and 'reference resolution'--before considering other topics that fit into the paradigm known as 'experimental pragmatics'.     

Appreciation of pragmatic interpretations of indirect commands: comparison of right and left hemisphere brain-damaged patients

Indirect commands can elicit "literal" interpretations, directly reflecting the content and form of the utterance, or "pragmatic" interpretations, incorporating knowledge of paralinguistic, social, or historical cues. The aim of this study was to demonstrate how right brain-damaged (RBD) subjects would perform on judgments of these forms of communication. It was predicted that relative to normal controls or aphasic subjects, the RBD subjects would have significant difficulty appreciating the pragmatic interpretation relative to the literal one. Subjects judged the appropriateness of two-part exchanges consisting of indirect commands, direct commands, Wh-questions coupled with pragmatic responses, literal responses, responses verifying the physical surround, or syntactically similar responses. The results confirm that the RBD subjects had a selective difficulty appreciating the indirect commands. Their preference of literal over pragmatic interpretations was significantly different to that of aphasics or normal controls. Hypotheses are offered to explain these data.     

Impact of emotional content on discourse production in patients with unilateral brain damage.

A picture story task was developed to examine expression of emotion via the verbal/lexical channel. The task elicited discourse with either emotional, visual-spatial, or neutral content and was administered to right brain-damaged (RBD), left brain-damaged (LBD), and normal control (NC) right-handed adults. Subjects were matched for gender, age, education, and occupational status. The brain-damaged groups were matched for months post-CVA onset and were similar with respect to intrahemispheric site of lesion. While the number of words produced was equivalent for each of the subject groups, the RBDs and LBDs expressed quantitatively less content than did the NCs. When content differences were examined within each subject group, there were no differences for LBDs and NCs, but the RBDs showed a selective deficit when producing emotional content. This finding suggests a special role for the right hemisphere in the production of emotional content in verbal discourse.     

Speech-sounds perception test: long- and short-forms reliability adjusted for item difficulty

The internal consistency reliability of the Speech-Sounds Perception Test was computed for the long form (.9159), short form (.8410), and the six series (.6109 to .7310). The total sample (N= 632) consisted of normal, diffusely brain-damaged, and undiagnosed patients referred for assessment.     

Effects of right and left hemisphere damage on performance of the "Right Hemisphere Communication Battery"

A Hebrew adaptation of Gardner and Brownell's (1986) "Right Hemisphere Communication Battery" (HRHCB) was administered to 27 right brain-damaged (RBD) patients, 31 left brain-damaged (LBD) patients, and 21 age-matched normal controls. Both patient groups showed deficits relative to controls and overall there was no difference between the two patient groups. A factor analysis of patients' scores on the HRHCB yielded two interpretable factors, a verbal and a nonverbal one. These factors were not lateralized. Performance of patients on the HRHCB correlated significantly and positively with performance on most tests of basic language functions, measured with a Hebrew adaptation of the "Western Aphasia Battery" (HWAB) and with other cognitive functions measured with standardized neuropsychological tests. There were stronger correlations of HRHCB with subtests of the HWAB in LBD patients and with nonlanguage cognitive tests in RBD patients. In the LBD group, HRHCB subtests' scores correlated negatively with lesion extent in frontal and temporal perisylvian regions. Such localization was not observed in RBD patients. The results argue against selective right hemisphere (RH) involvement in the RHCB, alleged to measure pragmatic aspects of language use, and show, instead, bilateral involvement. The results also argue against a modular organization of these functions of language use, especially in the RH.     

Perception and production of tone in aphasia

An acoustical and perceptual study of lexical tone was conducted to evaluate the extent and nature of tonal disruption in aphasia. The language under investigation was Thai, a tone language which has five lexical tones--mid, low, falling, high, and rising. Subjects included six left brain-damaged aphasics (two Broca's, one transcortical motor, one global, one conduction, one Wernicke), one right brain-damaged nonaphasic, one cerebellar dysarthric, and five normals. High-quality tape recordings of each subject's productions of a minimal set of five, monosyllabic Thai words were presented to 10 adult Thai listeners for identification. Results from the phonemic identification tests indicated that tone production is relatively spared in aphasic patients with unilateral left hemisphere lesions. The performance of the global aphasic, however, was considerably below normal. Patterns of tonal confusions further revealed that the performance of all aphasics, except the global, differed from that of normal speakers primarily in degree rather than in kind. Tonal contrasts were signaled at a high level of proficiency by the right brain-damaged and dysarthric patients. Acoustical analysis revealed that F0 contours associated with the five tones for all aphasics, except the global, were similar in overall shape as well as position in the tone space to those of normals. F0 contours for the right brain-damaged patient and the dysarthric also generally agreed with those of normals in terms of shape and position. F0 ranges of both aphasic and nonaphasic brain-damaged speakers were generally larger than those of normals for all five tones. The relationship between tone and vowel duration was generally similar to that of normals for all brain-damaged speakers. A comparison of aphasics' performance on tone perception (J. Gandour & R. Dardarananda, 1983, Brain and Language, 18, 94-114) and tone production indicated that, for the normal and right brain-damaged subjects, performance on the perception task was higher than on production, whereas the opposite was true for the aphasics. These data are brought to bear on issues related to tone production in aphasia, consonant and vowel production in aphasia, hemispheric specialization for tone production, intonation production in aphasia, relationship between speech perception and speech production, and tone production in dysarthria with cerebellar disease.     

Perceiver and poser asymmetries in processing facial emotion

This study examined both perceiver and poser asymmetries in processing facial emotion. Posers were left brain-damaged (LBD), right brain-damaged (RBD), and normal control (NC) right-handed males videotaped while expressing happiness and anger. Perceivers rated the facial expressions for asymmetry in original and reversed orientations. Overall, expressions viewed in the reversed orientation were rated as more left-sided than in the original orientation. In the reversed orientation, the more extensive left hemiface of the NCs and LBDs fell in the perceiver's left hemispace. This finding is consistent with previous research demonstrating a left hemispace bias for free-field viewing of emotional faces. Expressions were produced significantly more intensely on the left than the right hemiface by NCs and LBDs; expressions of RBDs were not significantly lateralized.     

Cognitive discrimination in brain-damaged adults: color vs form preference

The relative preference of color, form, and other dimensions as the basis for discrimination between stimuli has been investigated in a variety of normal and language-impaired populations. In aphasic adults, previous studies have been contradictory, and no studies of non-aphasic, brain-damaged adults have been reported. In this study color vs form preference was examined in 20 left brain-damaged, aphasic adults, 20 right brain-damaged non-aphasic adults, and 20 control subjects. Significant differences were found among all groups, with preference of color over form greatest for left brain-damaged subjects, followed by right-damaged and control subjects.     

The perception of positive and negative facial expressions by unilateral stroke patients

There remains conflict in the literature about the lateralisation of affective face perception. Some studies have reported a right hemisphere advantage irrespective of valence, whereas others have found a left hemisphere advantage for positive, and a right hemisphere advantage for negative, emotion. Differences in injury aetiology and chronicity, proportion of male participants, participant age, and the number of emotions used within a perception task may contribute to these contradictory findings. The present study therefore controlled and/or directly examined the influence of these possible moderators. Right brain-damaged (RBD; n = 17), left brain-damaged (LBD; n = 17), and healthy control (HC; n = 34) participants completed two face perception tasks (identification and discrimination). No group differences in facial expression perception according to valence were found. Across emotions, the RBD group was less accurate than the HC group, however RBD and LBD group performance did not differ. The lack of difference between RBD and LBD groups indicates that both hemispheres are involved in positive and negative expression perception. The inclusion of older adults and the well-defined chronicity range of the brain-damaged participants may have moderated these findings. Participant sex and general face perception ability did not influence performance. Furthermore, while the RBD group was less accurate than the LBD group when the identification task tested two emotions, performance of the two groups was indistinguishable when the number of emotions increased (four or six). This suggests that task demand moderates a study’s ability to find hemispheric differences in the perception of facial emotion.     

Hemispheric asymmetry in a perceptual priming task: evidence from patients with unilateral brain damage.

The present study examined the relationship between unilateral brain damage and the effect of nonverbal perceptual priming using a picture-fragment completion task. Subjects consisted of 11 left brain-damaged patients, 8 right brain-damaged patients, and 10 healthy normal controls. The mean score of normal controls was significantly higher than those of left and right brain-damaged patients. Although there were significant effects of priming in all the groups, a significant difference in amplitude of priming effects was found only between right brain-damaged patients and normal controls. The correlation between amount of fragmentation at which there was identification and priming effect was not significant. We conclude that the right hemisphere is more involved in perceptual priming of form than is the left hemisphere, and form-dependent processing in the perceptual priming task has an asymmetrical distribution in the right and left hemispheres.     

Effects of left-hand preference on postinjury measures of distal motor ability

The relationship of preinjury left-hand dominance for motor performance to postinjury distal motor skills following penetrating brain wounds in patients without overt hemiplegia was examined. We studied 13 controls, 13 right-hemisphere brain-damaged patients, and 11 left-hemisphere brain-damaged patients on motor tasks measuring reaction time, strength, and coordination. Our results indicated that no persistent deficits were seen on distal motor tasks in left-handed adults who suffered a penetrating brain wound. These findings are compatible with the relative sparing of persistent neuropsychological deficits in left-handers following brain damage.     

The Use of Contextual Information Related to General World Knowledge by Right Brain-Damaged Individuals in Pronoun Resolution

This study investigated the ability of right brain-damaged individuals (RBD) to use contextual information to resolve ambiguous pronouns. Subjects were presented with sentence pairs and required to resolve the ambiguous pronoun in the second sentence. Contrary to the prevailing view that RBD patients have difficulty using contextual information to integrate language, the RBD group demonstrated a normal pattern of response, demonstrating a sensitivity to the pragmatic information contained in the leading sentence. They responded more quickly to sentences with a pragmatically constrained preferred referent than to those sentences for which there was no preferred referent. As well, they chose the preferred referent significantly more often than the non-preferred referent. These results suggest that RBD patients can use contextual information at the level of a minimal discourse (i.e., two sentences).     

The validity of a computerized visual searching task as an indicator of brain damage

Goldstein, Rennick, Welch, and Shelly (1973) developed a visual searching task (VST) that succeeded in obtaining a hit rate of 94.1% correct classifications when comparing brain-damaged and normal subjects and a 79.4% hit rate when comparing brain-damaged and psychiatric subjects. Goldstein and Kyc (1978) reported 92.5% correct classifications for the brain-damaged vs. normal comparisons and 82.5% correct classifications for brain-damaged vs. schizophrenic comparisons. We computerized the administration of the VST and found 85.7% correct classifications for the brain-damaged vs. normal groups and 71.4% correct classifications for the brain-damaged vs. psychiatric group. These results suggest that the computerized VST (CVST) is also a potentially valid indicator of brain damage.     

Création et validation d’une version française du questionnaire AttrakDiff pour l’évaluation de l’expérience utilisateur des systèmes interactifs

IntroductionWhile user experience (UX) evaluation is a core concern within the field of human–computer interaction (HCI), there is currently no valid self-administered UX evaluation tool in French. The AttrakDiff 2 scale (Hassenzahl, Burmester, & Koller, 2003) is a UX evaluation tool, which relies on a theoretical model distinguishing pragmatic and hedonic qualities of interactive systems.ObjectiveThis paper describes the translation and validation of the French version of the AttrakDiff 2 scale in order to ease UX assessment in French-speaking users.MethodFollowing the cross-cultural methodology developed by Vallerand (1989), the questionnaire was translated by trilingual researchers before being back-translated and validated by a panel of experts. A pre-test was conducted on 26 participants. The characteristics of the French version of the AttrakDiff 2 scale were then evaluated through a quantitative online study involving a sample of 381 users.ResultsThe results confirm the expected 3 factors structure and a good internal consistency of each subscale. The links between factors are consistent with Hassenzahl's theoretical model (2003) where pragmatic and hedonic perceived attributes combine to form a judgment of attractiveness.ConclusionThe current French version of the AttrakDiff 2 scale is globally reliable with regards to the initial German version and presents satisfactory levels of validity and reliability.     

Emotion recognition in stroke patients with left and right hemispheric lesion: results with a new instrument-the FEEL Test

The aim of this study was to investigate the effect of a stroke event on people's ability to recognize basic emotions. In particular, the hypothesis that right brain-damaged (RBD) patients would show less of emotion recognition ability compared with left brain-damaged (LBD) patients and healthy controls, was tested. To investigate this the FEEL Test (Facially Expressed Emotion Labeling) was used, a computer based psychometric test that assesses one's ability to recognize facially displayed basic emotions via a forced-choice paradigm. We examined 24 patients after a stroke event (13 RBD, 11 LBD) and compared them with a matched group of healthy controls (HC, n=29). Results showed that the stroke patients performed significantly worse in the FEEL Test than did HC (p<.001). This deficit was especially evident for negative emotions (fear, anger, sadness, and disgust). In contrast to other studies we did not find any significant differences between RBD and LBD patients in their ability to recognize emotions. These results indicate that a stroke event has a negative effect on the recognition of facially displayed emotions but suggest that this effect is apparently not dependent on the side of the brain damage.     

Are brain-damaged patients inappropriately labeled as malingering using the 21-item Test and the WMS-R Logical Memory Forced Choice Recognition Test

This study examined the false positive hit rate of the 21-item Test and the WMS-R Logical Memory Forced Choice Recognition Test and compared the relationship between the measures in classifying biased responding/malingering. Of 40 patients referred for comprehensive neuropsychological evaluation, 18 were classified as brain-damaged based on independent neurological or neuroradiological examination. The remaining 22 patients could not be so classified on neurological or neuroradiological evidence and thus served as a medical control group. Findings indicated the brain-damaged group performed more poorly than did the control group across measures, and both groups, on the average, performed markedly better than that required to suggest biased responding. Also, taking both tests together, no individual patient was classified as malingering.     

Inference of object use from pantomimed actions by aphasics and patients with right hemisphere lesions

Twenty-four aphasic and fifteen right brain-damaged subjects were compared on their ability to identify the objects whose use was depicted in a series of twenty videotaped pantomimes. Aphasics were inferior to right brain-damaged patients in inferring object use. Success was correlated with Performance IQ, but not with language measures. Analysis of movement features contributing to subjects' choices reveal speed of movement and object weight to be the most robust and hand shape and size to be the most fragile.This study was supported in part by the grants from the National Institute of Health (RO1 EY07861) and McDonnell Pew Foundation (T89-01245-017) to LMV and by the USPHS grant NS06209 to H.G. and L.M.V. The authors are indebted to the staff of the Neuropsychology and Communication Disorders Departments at the New England Rehabilitation Hospital for making available the test records of the patients included in this study. All the patients included in the study have signed the informed consent forms according to the regulations of the Human Subjects Committee at New England Rehabilitation Hospital.     

Development of procedures for rating posed emotional expressions across facial, prosodic, and lexical channels.

A number of rating systems are available to evaluate emotional communication in a single modality. The main purpose of this study was to develop procedures to train human raters to evaluate posed expressions of emotion across three different channels of communication, i.e., facial, prosodic/intonational, and lexical/verbal. These procedures were used to evaluate posed emotional expressions produced by individuals with unilateral brain lesions from stroke. Posers in this preliminary report were two right brain-damaged, two left brain-damaged, and two normal control right-handed adults who were matched on demographic and neurological factors. Eight emotional expressions, both positive and negative, were produced in three channels and rated for intensity, pleasantness, and category accuracy. 15 normal adults served as raters, five per channel. The rating procedures were comparable across channels, with analogous properties, and yielded substantial interrater agreement. In this small sample of posers, it was observed that the expressions of the right brain-damaged group were rated as the least accurate and those of the left brain-damaged group as the most intense. When patterns of individual performance across the channels were examined, performance was quiet consistent for the normal controls yet variable for the right brain-damaged persons. These observations are in keeping with the notion that patients with right hemisphere pathology have difficulty in emotional communication. In summary, these findings suggest that comparison of emotional expressions across multiple channels is feasible.     

Text comprehension in aphasia

The semantic and pragmatic strategies in the comprehension of spoken texts are investigated in four subgroups of aphasic patients and in normal and brain-damaged controls. Short texts of similar linguistic structure were read to the subjects, who were required to choose the picture, from a multiple choice set of five, that was appropriate to the story. Besides a picture showing the main event of the story, one picture depicted the literal sense of a metaphorical comment, and the others misrepresented semantic functions expressed in the text. With respect to these types of responses, both aphasics and controls exhibited the same pattern of reaction. It is concluded that the redundancy of texts makes up for the difficulties aphasics have in comprehending isolated words and sentences.