Abstractive knowledge

 

The effect of concreteness

- In the article Semantic memory >> we primarily focused on the physical properties of concrete objects. Clearly, a complete theory of semantic memory must also provide an explanation of how we represent abstract concepts (e.g., peace, happiness, success) as well as abstract properties of concrete objects (e.g., "used to tell time" is a property of a clock). . According to the “concreteness effect,” concrete words should be processed more easily than abstract words because their representations include sensory information, which abstract words lack.

However, there have been reports of patients with semantic dementia having more difficulty with concrete than abstract words, suggesting that the difference between these types of words must be more than the amount of information.

One piece of evidence for qualitative differences between representations of concrete and abstract words comes from the work of Crutch and Warrington (2005). An example of an AZ patient with left temporal, parietal and posterior frontal damage. For concrete words, she had more interference with words that are closely related in meaning (eg, synonyms) than with "related" words (ie, words that share minimal meaning but often occur in similar contexts). For abstract words, interference was the opposite.

Abstract connection

Neuroimaging studies comparing abstract and concrete words have identified an inconsistent array of regions associated with abstract concepts in the left superior temporal lobe (Wise et al, 2000), the left posterior middle temporal lobe (Grossman et al, 2002), or the right anterior temporal pole. .

These inconsistencies may be due to the different testing approaches used in these studies or to differences in the way "abstract" is framed. The operational definition of abstract may be particularly important because it varies greatly across studies – from words without sensorimotor associations to words that are not pictorial (ie, difficult to visualize) to emotion words (eg, love). We hypothesize that these differences are likely to have a particularly significant impact on where brain activation is observed.

Using abstract stimuli that have minimal sensorimotor associations, Noppeney and Price (2004) observed fMRI activation while subjects reasoned about words (nouns, verbs, and adjectives) related to visual, auditory, “tool,” and abstract concepts. Compared to others,

abstract words activated classical language regions: left inferior frontal lobe, left middle temporal lobe, left superior temporal sulcus, and left anterior temporal pole.

Since these are classic "language" areas, the authors suggest that the activations are due to abstract word representations relying more on contextual information provided by language.

Also, Rodriguez and colleagues (2011) confirmed activation in the same regions for abstract verbs and added that:

Abstract verbs activate more regions than concrete verbs.

All led to the hypothesis that these abstract networks are more widespread, due to the different contexts in which abstract words can be found (Hoffman et al, 2011). Like abstract words, abstract features (eg, "used to tell the time") have no direct sensorimotor correlates.

Our ability to imagine abstract concepts and features—that is, knowledge that cannot be directly perceived from any individual sensory modality—shows that semantic knowledge must be more than simple sensorimotor resonance.

How might abstract concepts or features be represented in the distributed architecture we hypothesized?

Roger and colleagues articulated a model of semantic memory that includes units that integrate information across attribute domains (including verbal descriptions and object names). As a consequence, "abstract semantic representations appear as a product of statistical learning mechanisms in a region of the cortex that is suitable for performing cross-modal mappings due to its numerous interconnections with different perceptual-motor areas" (Rogers et al., 2004).

The process of abstraction from modality-specific representations may occur gradually across a range of cortical regions (perhaps converging at the temporal pole). As a result, an abstraction gradient may appear in representations across a given cortical region (eg, the ventral extrastriate visual pathway), and the anterior shift may reflect the activation of a more abstract representation (Kosslin & Thompson, 2000). In other words, the conceptual space of similarity in more frontal regions may move slightly away from the space of similarity in the environment, moving in the direction of abstract relations.

„Interakcija između sadržaja koji nosi perceptualne reprezentacije i verbalnih oznaka proizvodi prostor sličnosti koji nije obuhvaćen nijednim domenom pojedinačnih atributa, već odražava apstraktnu sličnost (Caramazza, Hillis, Rapp, & Romani, 1990; Chatterjee, 2010; Damasio, 1989; Plaut, 2002; Tiler, Moss, Durrant-Peatfield, & Levi, 2000).

"The interaction between content carrying perceptual representations and verbal labels produces a space of similarity that is not encompassed by any domain of individual attributes, but rather reflects abstract similarity (Caramazza, Hillis, Rapp, & Romani, 1990; Chatterjee, 2010; Damasio, 1989; Plaut, 2002; Tiller, Moss, Durrant-Peatfield, & Levi, 2000).

Abstraction and semantics

For this passage it is good to return to: Semantic memory >>

Based on the abundant interconnections between the temporal pole and various sensorimotor areas and the fact that temporal pole degeneration is associated with semantic dementia, Rogers and colleagues (2004) suggested that this region may support abstract knowledge and generalization.

Semantic dementia in particular has had a major impact on ideas about the role of the anterior temporal lobes in semantic memory. In this disorder, relatively focal (localized) degeneration in the anterior temporal lobes is accompanied by a lack of semantic memory (eg, problems in naming, recognizing and classifying objects, regardless of category), while other cognitive functions are relatively spared (Hodges & Patterson, 2007). .

The combination of anatomical and cognitive impairments in semantic dementia therefore lends credence to the idea that the anterior temporal lobes are important for supporting semantic memory, abstract knowledge, and generalization abilities.

Finally, it remains an open question whether regions outside the anterior temporal lobe serve similar "converging" functions.

Reference

- Caramazza, A., Hillis, A. E., Rapp, B. C., & Romani, C. (1990). The multiple semantics hypothesis: Multiple confusions? Cognitive Neuropsychology, 7, 161–189

- Chatterjee, A. (2010). Disembodying cognition. Language and Cognition. 2-1, 79–116

- Crutch, S. J., & Warrington, E. K. (2005). Abstract and concrete concepts have structurally different representational frameworks. Brain, 128, 615–627.

- Grossman, M., Koenig, P., DeVita, C., Glosser, G., Alsop, D., Detre, J., & Gee, J. (2002). The neural basis for category-specific knowledge: An fMRI study. NeuroImage, 15, 936–948.

- Hodges, J. R., & Patterson, K. (2007). Semantic dementia: a unique clinicopathological syndrome. The Lancet Neurology, 6, 1004–1014.

- Hoffman, P., Rogers, T. T., & Lambon Ralph, M. A. (2011). Semantic diversity accounts for the “missing” word frequency effect in stroke aphasia: Insights using a novel method to quantify contextual variability in meaning. Journal of Cognitive Neuroscience, 23, 2432–2446.

- Kosslyn, S. M., & Thompson, W. L. (2000). Shared mechanisms in visual imagery and visual perception: Insights from cognitive neuroscience. In M. S. Gazzaniga (Ed.), The new cognitive neurosciences (2nd ed., pp. 975–985). Cambridge, MA: MIT Press.

- Noppeney, U., & Price, C. J. (2004). Retrieval of abstract semantics. NeuroImage, 22, 164–170.

- Rogers, T. T., Lambon Ralph, M. A., Garrard, P., Bozeat, S., McClelland, J. L., Hodges, J. R., & Patterson, K. (2004). The structure and deterioration of semantic memory: A neuropsychological and computational investigation. Psychological Review, 111, 205–235.

- Rodriguez-Ferreiro, J., Gennari, S. P., Davies, R., & Cuetos, F. (2011). Neural correlates of abstract verb processing, Journal of Cognitive Neuroscience, 23, 106–118.

- Wise, R. J. S., Howard, D., Mummery, C. J., Fletcher, P., Leff, A., Buchel, C., & Scott, S. K.(2000). Noun imageability and the temporal lobes. Neuropsychologia, 38, 985–994.