The representation of ontological category concepts as affected by healthy aging: normative data and theoretical implications

The impact of ChatGPT on human data collection: A case study involving typicality norming data

Tools like ChatGPT, which allow people to unlock the potential of large language models (LLMs), have taken the world by storm. ChatGPT's ability to produce written output of remarkable quality has inspired, or forced, academics to consider its consequences for both research and education. In particular, the question of what constitutes authorship, and how to evaluate (scientific) contributions has received a lot of attention. However, its impact on (online) human data collection has mostly flown under the radar. The current paper examines how ChatGPT can be (mis)used in the context of generating norming data. We found that ChatGPT is able to produce sensible output, resembling that of human participants, for a typicality rating task. Moreover, the test-retest reliability of ChatGPT's ratings was similar to that of human participants tested 1 day apart. We discuss the relevance of these findings in the context of (online) human data collection, focusing both on opportunities (e.g., (risk-)free pilot data) and challenges (e.g., data fabrication).

Category production norms for 117 concrete and abstract categories

We present a database of category production (aka semantic fluency) norms collected in the UK for 117 categories (67 concrete and 50 abstract). Participants verbally named as many category members as possible within 60 seconds, resulting in a large variety of over 2000 generated member concepts. The norms feature common measures of category production (production frequency, mean ordinal rank, first-rank frequency), as well as response times for all first-named category members, and typicality ratings collected from a separate participant sample. We provide two versions of the dataset: a referential version that groups together responses that relate to the same referent (e.g., hippo, hippopotamus) and a full version that retains all original responses to enable future lexical analysis. Correlational analyses with previous norms from the USA and UK demonstrate both consistencies and differences in English-language norms over time and between geographical regions. Further exploration of the norms reveals a number of structural and psycholinguistic differences between abstract and concrete categories. The data and analyses will be of use in the fields of cognitive psychology, neuropsychology, psycholinguistics, and cognitive modelling, and to any researchers interested in semantic category structure. All data, including original participant recordings, are available at https://osf.io/jgcu6/ .

A category-selective semantic memory deficit for animate objects in semantic variant primary progressive aphasia

Data are mixed on whether patients with semantic variant primary progressive aphasia exhibit a category-selective semantic deficit for animate objects. Moreover, there is little consensus regarding the neural substrates of this category-selective semantic deficit, though prior literature has suggested that the perirhinal cortex and the lateral posterior fusiform gyrus may support semantic memory functions important for processing animate objects. In this study, we investigated whether patients with semantic variant primary progressive aphasia exhibited a category-selective semantic deficit for animate objects in a word-picture matching task, controlling for psycholinguistic features of the stimuli, including frequency, familiarity, typicality and age of acquisition. We investigated the neural bases of this category selectivity by examining its relationship with cortical atrophy in two primary regions of interest: bilateral perirhinal cortex and lateral posterior fusiform gyri. We analysed data from 20 patients with semantic variant primary progressive aphasia (mean age = 64 years, S.D. = 6.94). For each participant, we calculated an animacy index score to denote the magnitude of the category-selective semantic deficit for animate objects. Multivariate regression analysis revealed a main effect of animacy (β = 0.52, t = 4.03, P < 0.001) even after including all psycholinguistic variables in the model, such that animate objects were less likely to be identified correctly relative to inanimate objects. Inspection of each individual patient's data indicated the presence of a disproportionate impairment in animate objects in most patients. A linear regression analysis revealed a relationship between the right perirhinal cortex thickness and animacy index scores (β = -0.57, t = -2.74, P = 0.015) such that patients who were more disproportionally impaired for animate relative to inanimate objects exhibited thinner right perirhinal cortex. A vertex-wise general linear model analysis restricted to the temporal lobes revealed additional associations between positive animacy index scores (i.e. a disproportionately poorer performance on animate objects) and cortical atrophy in the right perirhinal and entorhinal cortex, superior, middle, and inferior temporal gyri, and the anterior fusiform gyrus, as well as the left anterior fusiform gyrus. Taken together, our results indicate that a category-selective semantic deficit for animate objects is a characteristic feature of semantic variant primary progressive aphasia that is detectable in most individuals. Our imaging findings provide further support for the role of the right perirhinal cortex and other temporal lobe regions in the semantic processing of animate objects.

Age-Related Degree and Criteria Differences in Semantic Categorization

Individual differences in semantic categorization are commonplace. Individuals apply a word like SPORTS to different instances because they employ different conditions for category membership (vagueness in criteria) or because they differ regarding the extent to which they feel the term can be applied given fixed conditions (vagueness in degree). Three individuals may, for instance, disagree as to whether chess and hiking are SPORTS, because one believes SPORTS are competitive in nature, while the other two require SPORTS to be effortful (vagueness in criteria). On the basis of whether they consider hiking sufficiently effortful or not, the latter two individuals might still disagree as to whether to call it a SPORT (vagueness in degree). We investigated whether there are systematic age-related differences in semantic categorization by analyzing the categorization decisions of 1,868 adults for eight semantic categories with a formal model that allows the two sources of categorization differences to be disentangled. We found that young and older adults assess instances differently with respect to the categorization conditions and that older adults employ a lower threshold for category membership than young adults do. We recommend that these criteria and degree differences are taken into account in studies of age-related semantic processing.

Can prediction-based distributional semantic models predict typicality?

Recent advances in the field of computational linguistics have led to the development of various prediction-based models of semantics. These models seek to infer word representations from large text collections by predicting target words from neighbouring words (or vice versa). The resulting representations are vectors in a continuous space, collectively called word embeddings. Although psychological plausibility was not a primary concern for the developers of predictive models, it has been the topic of several recent studies in the field of psycholinguistics. That is, word embeddings have been linked to similarity ratings, word associations, semantic priming, word recognition latencies, and so on. Here, we build on this work by investigating category structure. Throughout seven experiments, we sought to predict human typicality judgements from two languages, Dutch and English, using different semantic spaces. More specifically, we extracted a number of predictor variables, and evaluated how well they could capture the typicality gradient of common categories (e.g., birds, fruit, vehicles, etc.). Overall, the performance of predictive models was rather modest and did not compare favourably with that of an older count-based model. These results are somewhat disappointing given the enthusiasm surrounding predictive models. Possible explanations and future directions are discussed.

Acquiring New Factual Information: Effect of Prior Knowledge

One influential theory on object knowledge is feature-based model, which proposes that the object knowledge is organized by different feature types, such as sensory/perceptual and motor/functional ones. Previous studies have shown that prior knowledge enhances the processes of acquiring and remembering relevant information. However, whether the effect of prior knowledge is applied to different types of conceptual information over time remains unclear. In this study, we addressed this question by testing memory of different types of object features at various retention intervals. The level of prior knowledge was manipulated as object features from familiar and unfamiliar categories. In Experiments 1 and 2, sentences that described the perceptual and functional features of new words were presented. Sentences with episodic features were additionally presented in Experiment 2. The participants were then tested with recognition (Experiment 1) and recall (Experiment 2) tasks at different retention intervals. The results showed that prior knowledge enhanced memory for perceptual features but not for functional and episodic features. Such enhancement depended on the recollection process. In addition, the effect of prior knowledge on perceptual features remained stable over time. This study clarified how different types of new factual information were acquired and maintained and highlighted the importance of prior knowledge in acquiring new conceptual knowledge with the passage of time.

Word Familiarity Modulated the Effects of Category Familiarity on Memory Performance

Previous studies have shown that prior knowledge can have both enhancing and detrimental effects on memory for relevant information. Few studies have explored the boundary conditions under which prior knowledge facilitates or interferes with memory processes. In addition, to what extent the effects of prior knowledge change over time is unclear. In this study, we addressed this question by separating category familiarity (i.e., prior conceptual knowledge) and stimulus familiarity at different retention intervals. Participants were tested with a recognition task after they learned four types of words, that is., familiar words from familiar categories (FwordFcate) and unfamiliar categories (FwordUcate) as well as unfamiliar words from familiar (UwordFcate) and unfamiliar categories (UwordUcate). The results showed a significant interaction between category familiarity and word familiarity, that is, unfamiliar words, but not familiar words, from familiar categories were remembered better than those from unfamiliar categories. The enhancing effect of category familiarity depended on the recollection process and remained stable over time. This study suggested that stimulus familiarity modulates the effects of category familiarity on memory performance, and clarified the boundary conditions for the effects of prior knowledge.

Vocabulary relearning in semantic dementia: Positive and negative consequences of increasing variability in the learning experience

Anomia therapy typically aims to improve patients' communication ability through targeted practice in naming a set of particular items. For such interventions to be of maximum benefit, the use of trained (or relearned) vocabulary must generalise from the therapy setting into novel situations. We investigated relearning in three patients with semantic dementia, a condition that has been associated with poor generalisation of relearned vocabulary. We tested two manipulations designed to improve generalisation of relearned words by introducing greater variation into the learning experience. In the first study, we found that trained items were retained more successfully when they were presented in a variety of different sequences during learning. In the second study, we found that training items using a range of different pictured exemplars improved the patients' ability to generalise words to novel instances of the same object. However, in one patient this came at the cost of inappropriate over-generalisations, in which trained words were incorrectly used to name semantically or visually similar objects. We propose that more variable learning experiences benefit patients because they shift responsibility for learning away from the inflexible hippocampal learning system and towards the semantic system. The success of this approach therefore depends critically on the integrity of the semantic representations of the items being trained. Patients with naming impairments in the context of relatively mild comprehension deficits are most likely to benefit from this approach, while avoiding the negative consequences of over-generalisation.

•

SD patients attempted to recover lost vocabulary by naming pictures daily.

•

Learning was more successful when pictures presented in a different order each day.

•

One patient benefitted from naming multipleexemplars of each item.

•

But in the other patient, this resulted in over-generalisation of the trained name.

Performance of younger and older adults on tests of word knowledge and word retrieval: independence or interdependence of skills?

PURPOSE

The purpose of this study was to examine the associations between vocabulary knowledge and word retrieval in younger and older adults.

METHOD

Three tests of word retrieval and 2 tests of word knowledge were administered to 140 Hebrew-speaking adults, half of whom were younger (M(age) = 24.20 years) and half of whom were older (M(age) = 74.83 years).

RESULTS

Younger adults outperformed older adults on tests of retrieval, whereas older adults outperformed younger adults on tests of vocabulary, and no association was found between the 2 skills across the entire sample. Once age and education were taken into account, both skills contributed to the prediction of each other and were similarly related within each group. Older adults performed equally well when required to produce and recognize word meanings, whereas younger adults were better at recognition than at production.

CONCLUSIONS

Older age is associated with better knowledge and with retrieval difficulties, yet individual differences in vocabulary within each age group affect level of retrieval, and variability in search skills affects performance on vocabulary tests. Although the assessment of vocabulary is not free of retrieval demands, older adults as a group are more successful than are younger adults at producing word definitions, most likely because their knowledge is more complete.

Apples are not the only fruit: the effects of concept typicality on semantic representation in the anterior temporal lobe

Intuitively, an apple seems a fairly good example of a fruit, whereas an avocado seems less so. The extent to which an exemplar is representative of its category, referred to here as concept typicality, has long been thought to be a key dimension determining semantic representation. Concept typicality is, however, correlated with a number of other variables, in particular age of acquisition (AoA) and name frequency. Consideration of picture naming accuracy from a large case-series of semantic dementia (SD) patients demonstrated strong effects of concept typicality that were maximal in the moderately impaired patients, over and above the impact of AoA and name frequency. Induction of a temporary virtual lesion to the left anterior temporal lobe, the region most commonly affected in SD, via repetitive Transcranial Magnetic Stimulation produced an enhanced effect of concept typicality in the picture naming of normal participants, but did not affect the magnitude of the AoA or name frequency effects. These results indicate that concept typicality exerts its influence on semantic representations themselves, as opposed to the strength of connections outside the semantic system. To date, there has been little direct exploration of the dimension of concept typicality within connectionist models of intact and impaired conceptual representation, and these findings provide a target for future computational simulation.

EEG decoding of semantic category reveals distributed representations for single concepts

Achieving a clearer picture of categorial distinctions in the brain is essential for our understanding of the conceptual lexicon, but much more fine-grained investigations are required in order for this evidence to contribute to lexical research. Here we present a collection of advanced data-mining techniques that allows the category of individual concepts to be decoded from single trials of EEG data. Neural activity was recorded while participants silently named images of mammals and tools, and category could be detected in single trials with an accuracy well above chance, both when considering data from single participants, and when group-training across participants. By aggregating across all trials, single concepts could be correctly assigned to their category with an accuracy of 98%. The pattern of classifications made by the algorithm confirmed that the neural patterns identified are due to conceptual category, and not any of a series of processing-related confounds. The time intervals, frequency bands and scalp locations that proved most informative for prediction permit physiological interpretation: the widespread activation shortly after appearance of the stimulus (from 100 ms) is consistent both with accounts of multi-pass processing, and distributed representations of categories. These methods provide an alternative to fMRI for fine-grained, large-scale investigations of the conceptual lexicon.

Explaining semantic short-term memory deficits: evidence for the critical role of semantic control

Patients with apparently selective short-term memory (STM) deficits for semantic information have played an important role in developing multi-store theories of STM and challenge the idea that verbal STM is supported by maintaining activation in the language system. We propose that semantic STM deficits are not as selective as previously thought and can occur as a result of mild disruption to semantic control processes, i.e., mechanisms that bias semantic processing towards task-relevant aspects of knowledge and away from irrelevant information. We tested three semantic STM patients with tasks that tapped four aspects of semantic control: (i) resolving ambiguity between word meanings, (ii) sensitivity to cues, (iii) ignoring irrelevant information and (iv) detecting weak semantic associations. All were impaired in conditions requiring more semantic control, irrespective of the STM demands of the task, suggesting a mild, but task-general, deficit in regulating semantic knowledge. This mild deficit has a disproportionate effect on STM tasks because they have high intrinsic control demands: in STM tasks, control is required to keep information active when it is no longer available in the environment and to manage competition between items held in memory simultaneously. By re-interpreting the core deficit in semantic STM patients in this way, we are able to explain their apparently selective impairment without the need for a specialised STM store. Instead, we argue that semantic STM patients occupy the mildest end of spectrum of semantic control disorders.

Generalization and differentiation in semantic memory: insights from semantic dementia

According to many theories, semantic representations reflect the parallel activation of information coded across a distributed set of modality-specific association brain cortices. This view is challenged by the neurodegenerative condition known as semantic dementia (SD), in which relatively circumscribed, bilateral atrophy of the anterior temporal lobes results in selective degradation of core semantic knowledge, affecting all types of concept, irrespective of the modality of testing. Research on SD suggests a major revision in our understanding of the neural basis of semantic memory. Specifically, it is proposed that the anterior temporal lobes form amodal semantic representations through the distillation of the multimodal information that is projected to this region from the modality-specific association cortices. Although cross-indexing of modality-specific information could be achieved by a web of direct connections between pairs of these regions, amodal semantic representations enable semantic generalization and inference on the basis of conceptual structure rather than modality-specific features. As expected from this hypothesis, SD is characterized by impaired semantic generalization, both clinically and in formal assessment. The article describes a comprehensive array of under- and overgeneralization errors by patients with SD when engaged in receptive and expressive verbal and nonverbal tasks and everyday behaviors.