Accessibility statement

Experimental adventures in CReLLU-land: On how verbal labels promote smell category learning

Wednesday 3 June 2020, 1.00PM to 2.00pm

Speaker(s): Dr Norbert Vanek, Centre for Research in Language Learning and Use

Recent research shows that speakers of most languages find smells difficult to abstract and name. Can verbal labels enhance the human capacity to learn smell categories? Few studies have examined how verbal labelling might affect non-visual cognitive processes, and thus far very little is known about word-assisted odour category learning.

To address these gaps, I tested whether different types of training change learning gains in odour categorization. After four intensive days of training to categorize odours that were co-presented with arbitrary verbal labels, people who learned odour categories with odour-label pairs that were more consistent were significantly more accurate than people with  the same perceptual experience but who had odour-label pairs that were less consistent (Fig. 1). Both groups' accuracy scores improved, but the learning curves differed. Only the context of consistent linguistic cuing supported a steady increase in correct responses from the onset of training.

The benefit from the results is twofold. First, they demonstrate that new verbal cues undergo multisensory integration with co-presented odours and affect the capacity to categorize olfactory input. And second, they demonstrate that consistent language input aids olfactory perceptual processing and improves expertise in odour categories markedly faster than inconsistent input. These findings are striking in that they suggest that not just naming, but the underlying conceptual representation of odours may differ across cultures in accordance with the language people speak.

Fig 1. Experiment design. (a) Sniffin’ sticks were used as olfactory stimuli. They were presented in a quiet and well-ventilated room, on a static platform, using an odourless glove, with the tip of the pen held around 2 centimetres under the participant’s nose. (b) Bisyllabic pseudowords used as verbal stimuli were co-presented with the odours. For the training phase, pseudoword pairs were presented under a consistent (CONCORD) or an inconsistent (DISCORD) condition. Each smell was presented 16 times in total, always within the same triplet. To manipulate the strength of associations formed during training, the consistency of odour-pseudoword pairs was 13/16 in the CONCORD condition and 4/16 in the DISCORD condition (as shown for mint-odourwithin the mint-eucalyptus-grassodour triplet). (c) The odour similarity judgement test used the same task design as the first day of training, except that the test was performed without presentation of pseudowords.

Location: via Zoom, link to be confirmed