Our senses take in a large amount of information, information that sometimes is congruent across sensory modalities. Crossmodal correspondences are the study of how this information across modalities is integrated by the brain, across which dimensions the correspondences exists, and how it affect us. In the present paper four experiments were conducted, in which potential crossmodal correspondences between audition, vision and olfaction were investigated. It was hypothesized that crossmodal correspondences between olfaction, vision and audition exist along different dimensions. The results showed significant correlations between olfaction and audition when volume varies, i.e., a high volume is associated to a high concentration of an odor, and a low volume is associated to a low concentration of an odor, and vice versa. Furthermore, existing correspondences between vision and audition is reconfirmed. In conclusion, the results provide support to the notion that crossmodal correspondences exists between all sensory modalities, although along different dimensions.

No description available.

The term 'crossmodal correspondence' refers to the tendency for people to match information across the senses. In this thesis, the associations between taste/flavour (tastants and words) information with shapes and colours is investigated. Furthermore, such correspondences are addressed in the context of multisensory packaging design. The focus in this thesis is on the way in which taste/flavour information can be communicated by means of the visual elements of product packaging. Through a series of experiments, I demonstrate that people associate tastes and the roundness/angularity of shapes, and that taste quality, hedonics, and intensity influence such correspondences. However, packaging roundness/angularity does not seem to drive these associations. Additionally, I demonstrate that culture and context systematically influence colour/flavour associations. Importantly, the results reported in this thesis suggest that taste/shape correspondences can influence taste expectations as a function of the visual attributes of product packaging. The results reported here also reveal that colour can influence the classification of, and search for, flavour information on a product’s packaging. It turns out that the strength of the association between a flavour category and a colour is crucial to such an effect. The implications of these findings are discussed in light of the theories of crossmodal correspondences, its applications, and directions for future research.

The physical properties of the distal stimuli activating our senses are often correlated in nature; it would therefore be advantageous to exploit such correlations to better process sensory information. Stimulus correlations can be contingent and readily available to the senses (like the temporal correlation between mouth movements and vocal sounds in speech), or can be the results of the statistical co-occurrence of certain stimulus properties that can be learnt over time (like the relation between the frequency of acoustic resonance and the size of the resonator). Over the last century, a large body of research on multisensory processing has demonstrated the existence of compatibility effects between individual features of stimuli from different sensory modalities. Such compatibility effects, termed crossmodal correspondences, possibly reflect the internalization of the natural correlation between stimulus properties. The present dissertation assesses the effects of crossmodal correspondences on multisensory processing and reports a series of experiments demonstrating that crossmodal correspondences influence the processing rate of sensory information, distort perceptual experiences and lead to stronger multisensory integration. Moreover, a final experiment investigating the effects of contingent signals’ correlation on multisensory processing demonstrates the key role of temporal correlation in inferring whether two signals have a common physical cause or not (i.e., the correspondence problem). A Bayesian framework is proposed to interpret the present results whereby stimulus correlations, represented on the prior distribution of expected crossmodal co-occurrence, operate as cues to solve the correspondence problem.