Elsevier

Neuropsychologia

Volume 64, November 2014, Pages 271-281
Neuropsychologia

Abstract and concrete categories? Evidences from neurodegenerative diseases

https://doi.org/10.1016/j.neuropsychologia.2014.09.041Get rights and content

Highlights

  • Category effects were investigated in AD and sv-PPA patients.

  • An abstractness effect was found in sv-PPA patients.

  • A better performance for non living vs. living items was infrequently present in AD.

  • A selective preservation of emotion was observed in AD.

  • A selective impairment of social relations was observed in sv-PPA.

Abstract

We assessed the performance of patients with a diagnosis of Alzheimer׳s disease (AD) and of the semantic variant of primary progressive aphasia (sv-PPA) in a series of tasks involving both abstract and concrete stimuli, which were controlled for most of the variables that have been shown to affect performance on lexical-semantic tasks.

Our aims were to compare the patients׳ performance on abstract and concrete stimuli and to assess category-effects within the abstract and concrete domains. The results showed: (i) a better performance on abstract than concrete concepts in sv-PPA patients. (ii) Category-related effects in the abstract domain, with emotion concepts being preserved in AD and social relations being selectively impaired in sv-PPA. In addition, a living–non living dissociation may be (infrequently) observed in individual AD patients after controlling for an extensive set of potential confounds.

Thus, differences between and within the concrete or abstract domain may be present in patients with semantic memory disorders, mirroring the different brain regions involved by the different pathologies.

Introduction

A cardinal property of human cognition is the ability to represent knowledge about the world. This is commonly referred to as semantic memory (Tulving, 1972). Semantic memory is assumed to provide for the organization of the knowledge about the manifold entities (i.e., idea, zebra, hammer, love, challenge), whose conceptual representations may be supported by different types of information. How concepts are represented in semantic memory is a matter of debate. Categorization ability is a key principle, allowing not only to group concepts together according to specific types of information, but also to extend to novel entities the general knowledge about a category inferred from its members. How categorization occurs, how semantic categories are organized, which concepts are grouped together and on which basis, are questions still far from having an answer (Caramazza and Shelton, 1998, Martin, 2007, Patterson et al., 2007).

The observation of neurological patients showing selective impairments of conceptual knowledge for specific categories has played a major role in the development of this area of research. Dissociations have been reported between different domains of knowledge, such as abstract and concrete concepts (Warrington, 1975). Moreover, within the concrete domain, a number of dissociations have been described among distinct categories, involving living (animals, fruits and vegetables) and non living entities (tools, vehicles, furniture) (for a review see Capitani et al, 2003). Dissociations between domains or categories of knowledge have been reported in patients with Alzheimer׳s Disease (AD) and the semantic variant of Primary Progressive Aphasia (sv-PPA) (Gorno-Tempini et al., 2011), also known as Semantic Dementia (SD) (Snowden et al., 1989, Hodges et al., 1992). Notwithstanding the considerable number of studies, the results are far from consistent. In particular, the same dissociation (i.e., abstract vs. concrete or living vs. non living) may be reversed/different in these two pathologies, or may also present in opposite directions in patients affected by the same condition. Only a very limited number of studies have investigated the abstract/concrete dichotomy in patients with AD, reporting a poorer performance for abstract than for concrete words (Rissenberg and Glanzer, 1987, Fung et al., 2000, Yi et al., 2007), or no significant differences (Graham et al., 2004, Hsieh et al., 2012). In contrast, studies on single cases or on groups of patients affected by SD have reported better performance on abstract than concrete concepts (i.e., reversal of the concreteness effect) (Warrington, 1975, Breedin et al., 1994, Cipolotti and Warrington, 1995, Macoir, 2009, Reilly et al., 2007, Papagno et al., 2009; Bonner et al., 2009; Yi et al., 2007). Some authors have even proposed that preserved abstract word knowledge is a typical feature of SD (Grossman and Ash, 2004), and that this effect is related to the severity of the semantic impairment (Bonner et al., 2009). Other studies, however, highlighted an impairment for both word types (Hoffman and Lambon Ralph, 2011, Jefferies et al., 2009), and have concluded that the disproportionate impairment for concrete words is rare and may be attributed either to individual differences in premorbid abilities (Jefferies et al., 2011) or to methodological confounds (i.e., higher familiarity for abstract than concrete concepts, or low imageability difference between abstract and concrete concepts) (Hoffman et al., 2012).

Within the concrete domain, the living–non living dissociation in AD is generally characterized by a better performance on non living items (Chertkow and Bub, 1990, Garrard et al., 1998, Gonnerman et al., 1997, Silveri et al., 1991, Zannino et al., 2002), although there are also studies reporting the opposite dissociation (Gonnerman et al., 1997) or no significant differences (Tippett et al., 1996). Only a limited number of case studies showed a dissociation between living and non-living in sv-PPA (i.e., Barbarotto et al., 1995, Lambon Ralph et al., 2003, Lambon Ralph et al., 1998; Zannino et al., 2006a), as sv-PPA patients generally show a poor performance for both categories. These studies, however, are difficult to compare, due to differences in the characteristics of stimuli and in the tasks.

In considering these inconsistencies, it has also been argued that materials were insufficiently matched for familiarity, and word frequency (Funnel and Sheridan, 1992); visual complexity (Stewart et al., 1992) and age of acquisition (Silveri et al., 2002). In addition, other factors, such as name agreement (Albanese, 2007), emotional valence (Brousseau and Buchanan, 2004) and manipulability (Filliter et al., 2005), or differences based on the semantic features of the concepts, such as semantic relevance (Sartori and Lombardi, 2004) and semantic distance (Zannino et al., 2006b) have also been shown to play a role. When categories were matched for many of the aforementioned variables, some studies found a persisting category effect (Silveri et al., 2002, Zannino et al., 2002, Martinaud et al., 2009), while others reported an attenuation or cancellation of the category effect (Tippett et al., 1996, Tippett et al., 2007). It is noteworthy that some investigators (Moreno-Martinez and Laws, 2008, Gale et al., 2009) have shown that AD performance on lexico-semantic tasks is only quantitatively worse, rather than qualitatively different from that of healthy elderly controls, with the same size and direction of the category effects.

Within the abstract domain, potential differences across categories have been seldomly investigated. Some patient studies have assumed that different organizational principles underlie the representation of concrete and abstract concepts (i.e., associative for abstract rather and categorical for concrete) (see Crutch and Warrington, 2005). This assumption, however, does not imply that different categories cannot be identified within the abstract domain, even if a greater overlap between abstract categories may exist, given the absence of distinct physical identifiers for category members. Nonetheless, in the few studies of AD and sv-PPA patients, abstract concepts have been grouped in a single category (Rissenberg and Glanzer, 1987, Jefferies et al., 2011). In one of the first reports concerning semantic memory impairments in AD, Martin and Fedio (1983) showed preserved word comprehension for emotion words. Word comprehension was assessed using a Symbol Referent Test, where subjects were asked to match “abstract pictorial representations with printed words denoting objects, actions, emotion, and modifiers”. AD patients were impaired in the comprehension of single words related to objects (table), actions (to sit), or modifiers (strong), but not for emotion words (happy, sad, hungry, love). However, a recent study assessing the comprehension of emotion words as compared to other abstract words and concrete words in both AD and sv-PPA, using two word comprehension tasks, one to assess abstract and concrete items, the other emotions, failed to find support for selective preservation of emotion words (Hsieh et al., 2012).

In addition, there is some evidence for concepts referring to social behaviors to be distinct from other abstract concepts. A selective activation of the bilateral superior anterior temporal lobes was highlighted when participants were asked to judge the meaning relatedness of social concepts (i.e., honor) as compared to concepts describing animal functions (i.e., nutritious) (Zahn et al., 2007).

The general aim of the current study was to assess semantic knowledge impairments in patients affected by neurodegenerative diseases, i.e., patients with AD and sv-PPA and for both concrete and abstract concepts. Differences in performance between patient groups may reflect the severity of a lexical-semantic impairment, or the different pattern of anatomical involvement in the early stage of disease, leading to qualitative differences between domains or categories of knowledge.

We specifically address the abstract/concrete dissociation and the presence of category effects in the abstract and concrete domains, trying to overcome some of the limitations of the previous studies. To assess concrete and abstract concepts, we use two batteries of tests involving different semantic categories, respectively living and non living entities for concrete concepts and human action, cognition, attribute, emotion and social relation for abstract concepts. The tests are characterized by an extremely tight control of the main variables that could influence performance on the tasks. We used two different batteries of tests specifically tailored to concrete (Catricalà et al., 2013) or to abstract concepts (Della Rosa et al., 2014). The necessity to create ad-hoc instruments to identify impairments in each knowledge domain type of knowledge is due to the difference in terms of hypothesized semantic representations (see Crutch and Warrington, 2005) for each type of knowledge, and to the difficulty or impossibility to investigate concepts in either domain through the same type of material (i.e., pictures or words) or with the same tools. In particular, the use of a single task for both types of concepts can actually result in an advantage of one domain over the other (i.e., association tasks are more suitable for abstract than for concrete words; synonym judgment tests are more suitable for concrete than abstract, in agreement with the hypothesis of different underlying organizations, see Crutch and Warrington, 2005). To the best of our knowledge, no previous study has compared AD patients and sv-PPA patients with controls using the same series of tasks employing both abstract and concrete concepts, assessing both input and output modalities and controlling for a large number of variables that could account for the subjects performance.

The abstract/concrete dissociation remains poorly investigated in AD and very controversial in sv-PPA. The few studies on AD reported a concreteness effect or no differences between abstract and concrete concepts. In sv-PPA, inconsistent results have been attributed to methodological confounds. In the present study, we looked for genuine abstract/concrete dissociations beyond potential methodological constraints related to tasks or stimuli characteristics in each domain, trying to overcome possible artefactual influences resulting in a differential performance for concrete or abstract concepts.

As for category-related effects, their presence in the concrete domain has been extensively investigated in AD. However, the studies are very heterogeneous in the results and in the variables taken in consideration. Within the concrete domain, we aim at identifying a genuine category-effect (i.e., living–nonliving), when taking into account those confounding variables that may account for the dissociation. Only a few studies have investigated category-related effects for abstract concepts. In AD contrasting evidence has been reported for emotion words, but the severity of pathology may explain any divergence. Our prediction was that emotions, considered as an abstract category per se, could be spared at least in an early stages of AD, when the atrophy is localized in mesial temporal structures (see Giffard et al., 2009 for similar consideration). Sv-PPA may also show an impairment for specific categories of abstract concepts depending upon the nature of their contextual constraints (see Wiemer-Hastings et al., 2001), the strength of associations between them (see Crutch and Warrington, 2005) or on the degree to which their representation hinges on brain structures known to be affected by this pathology, such as the anterior temporal lobe in the case of social concepts (see Zahn et al., 2007).

Section snippets

Subjects

Twenty patients (7 males) were included in the study: 14 AD and 6 sv-PPA (see Table 1).

The diagnosis of probable AD was made according to the criteria developed by the National Institute of Neurological and Communicative Disorders and Stroke (NINCDS) and the Alzheimer׳s Disease and Related Disorders Association (ADRDA) (McKhann et al., 1984). The AD group included subjects with mild to moderate levels of dementia (MMSE range 16–25) (Mini Mental State Examination, Folstein et al., 1975). The

Group level analyses

We first assessed the performance of AD and sv-PPA with respect to controls on all tasks, in order to individuate pathological performance at group level. Rank-based inverse normal transformation (transformation to rankit scores) was used to convert all the subjects scores obtained on the six tasks employed (i.e., for concrete items: picture naming, naming on oral description, sentence verification; for abstract items: sentence completion, multiple choice verbal matching, association task).

Discussion

In this study we assess the performance of a group AD and sv-PPA patients in a series of comprehension and production tasks employing both abstract and concrete stimuli. The aim was to assess the presence of different category effects, i.e., the dissociation between abstract and concrete concepts, between living and non living entities and between different abstract categories. Our study goes beyond those already in the literature because we used the same tasks across patients groups, we

Conclusion

The results of the analysis at both group and individual level support the claim for a relatively preserved performance for abstract word processing in patients with the semantic variant of primary progressive aphasia, which cannot be attributed to confounding factors, such as familiarity of imageability. In addition, what appear to be genuine category effects may be present in patients with semantic memory disorders, in both the concrete and abstract domains. Separate areas of activation for

Acknowledgments

This study was supported by a European Union (FT6-2004-NEST-PATH) Grant (028714) to S.F.C. and G.V.

References (80)

  • K.R. Laws

    Illusions of normality: a methodological critique of category-specific naming

    Cortex

    (2005)
  • J. Macoir

    Is a plum a memory problem? Longitudinal study of the reversal of concreteness effect in a patient with semantic dementia

    Neuropsychologia

    (2009)
  • A. Martin et al.

    Word production and comprehension in Alzheimer׳s disease: the breakdown of semantic knowledge

    Brain Lang.

    (1983)
  • M. Milders et al.

    Differential deficits in expression recognition in gene-carriers and patients with Huntington׳s disease

    Neuropsychologia

    (2003)
  • F.J. Moreno-Martínez et al.

    Domain or not domain? That is the question: longitudinal semantic deterioration in Alzheimer׳s disease

    Brain Cognit.

    (2011)
  • C. Papagno et al.

    Reversed of the concreteness effect for nouns in a subject with semantic dementia

    Neuropsychologia

    (2009)
  • J. Reilly et al.

    A unitary semantics account of reverse concreteness effects in semantic dementia

    Brain Lang.

    (2007)
  • L.J. Tippett et al.

    The semantic memory impairment of Alzheimer׳s disease: category-specific?

    Cortex

    (1996)
  • L.J. Tippett et al.

    Category specific deficits in Alzheimer׳s disease: fact or artefact?

    Cortex

    (2007)
  • M.P. Viggiano et al.

    A new standardized set of ecological pictures for experimental and clinical research on visual object processing

    Cortex

    (2004)
  • C. Whatmough et al.

    The semantic category effect increases with worsening anomia in Alzheimer׳s type dementia

    Brain Lang.

    (2003)
  • G.D. Zannino et al.

    The role of semantic distance in category-specific impairments for living things: evidence from a case of semantic dementia

    Neuropsychologia

    (2006)
  • G.D. Zannino et al.

    Category-specific impairment in patients with Alzheimer׳s disease as a function of disease severity: a cross-sectional investigation

    Neuropsychologia

    (2002)
  • G.D. Zannino et al.

    (Category-specific) semantic deficits in Alzheimer׳s patients: the role of semantic distance

    Neuropsychologia

    (2006)
  • R. Adolphs et al.

    Impaired recognition of emotion in facial expressions following bilateral damage to the human amygdala

    Nature

    (1994)
  • J. Altarriba et al.

    Concreteness, context availability and imageability ratings and words associations for abstract, concrete and emotion words

    Behav. Res. Methods, Instrum. Comput.

    (1999)
  • R. Barbarotto et al.

    Slowly progressive semantic impairment with category specificity

    Neurocase

    (1995)
  • M.F. Bonner et al.

    Reversal of the concreteness effect in semantic dementia

    Cognit. Neuropsychol.

    (2009)
  • S.D. Breedin et al.

    Reversal of the concreteness effect in a patient with semantic dementia

    Cognit. Neuropsychol.

    (1994)
  • E. Capitani

    Statistical methods

  • E. Capitani et al.

    Composite neuropsychological batteries and demographic correction: standardization based on equivalent scores, with a review of published data

    J. Clin. Expl. Neuropsychol.

    (1997)
  • E. Capitani et al.

    What are the facts of semantic category-specific deficits? A critical review of the clinical evidence

    Cognit. Neuropsychol.

    (2003)
  • A. Caramazza et al.

    Domain-specific knowledge systems in the brain: the animate-inanimate distinction

    J. Cognit. Neurosci.

    (1998)
  • E. Catricalà et al.

    An Italian battery for the assessment of semantic memory disorders

    Neurol. Sci.

    (2013)
  • L.L. Chao et al.

    Attribute-based neural substrates in temporal cortex for perceiving and knowing about objects

    Nat. Neurosci.

    (1999)
  • H. Chertkow et al.

    Semantic memory loss in dementia of Alzheimer׳s type. What do various measures measure?

    Brain

    (1990)
  • L. Cipolotti et al.

    Semantic memory and reading abilities: a case report

    J. Int. Neuropsychol. Soc.

    (1995)
  • J.R. Crawford et al.

    Testing for suspected impairments and dissociations in single-case studies in neuropsychology: evaluation of alternatives using Monte Carlo simulations and revised tests for dissociations”

    Neuropsychology

    (2005)
  • S.J. Crutch et al.

    Partial knowledge of abstract words in patients with cortical degenerative conditions

    Neuropsychology

    (2006)
  • S.J. Crutch et al.

    Abstract and concrete concepts have structurally different representational frameworks

    Brain

    (2005)
  • Cited by (0)

    View full text