The posters will be available throughout the symposium to view at the attendees discretion with formal presentations in our dedicated poster and demonstration sessions on Thursday and Friday between 12:30-13:15. We hope conversations can carry on into lunch and beyond!
See our contribution page here for further information on talks from our esteemed speakers and our interactive sensory demonstrators here.
| Presenter | Title | Abstract |
|---|---|---|
| Benjy Barnett | Identifying content-invariant neural signatures of phenomenal magnitude | Some experiences are stronger than others: they have greater phenomenal magnitude. Although the phenomenal magnitude of experience is a familiar component of human consciousness, the computational scheme underpinning its encoding in the brain is unknown. In particular, it is unknown whether phenomenal magnitude is encoded in a “rich” manner, via the strengthening or broadcast of content-specific perceptual representations, or a “sparse” scheme, in which a content-invariant signal monitors the reliability or precision of first-order contents. In a reanalysis of existing MEG and fMRI data from two distinct studies, we operationalise phenomenal magnitude as subjective ratings of awareness and visibility to address this question. We find that phenomenal magnitude is associated with content-invariant neural signatures distributed across visual, parietal, and frontal cortices. Moreover, we show that these neural signatures exhibit similarities with other analogue magnitude codes such as number, with each rating being represented as being more similar to its neighbours, and less similar to more distant ratings. Our findings suggest a role for a content-invariant analogue magnitude code in determining the strength of perceptual experience. |
| Brayan Mauricio Rodríguez Rivera | Exploring the relationship between sound and olfactory perception beyond flavors: A review of experimental studies | The belief that certain music tones are congruently associated with certain fragrances can be traced back 150 years. Indeed, olfactory perception depends on multisensory context and expectation. However, studying how auditory parameters can modulate olfactory experiences is mainly framed in the context of flavors. This study provides an overview of studies published since 2012 focusing on sounds corresponding to olfactory experiences outside the food industry. An average of 90 participants joined these studies, mainly answering scales that measured emotional perception, crossmodal congruency, and olfactory intensity. ANOVA was the most common data analysis technique. Music and non-music sounds (i.e., commercial songs, atmospheric sounds, pure tones, tailor-made soundtracks, and voiceovers) were used based on their pitch and evoked emotions. Lemon, coffee, raspberry, and lavender were the most common aromas, selected based on emotional attributes, intensity, and crossmodal congruence. Results show 17 auditory-olfactory associations, connecting sounds (e.g., ambient sounds, songs), and auditory parameters (e.g., pitch), to olfactory experiences (e.g., coffee aroma, discomfort). For example, coffee aroma is associated with coffee advertisements, as well as birds and conversation sounds. Although these crossmodal interactions between sounds and aromas in non-food related experiences are less studied, they can be helpful for industries such as pharma and cosmetics. Future work should consider online experiments outside the food context for e-commerce purposes. |
| Hamed Karimipour | Sensory representation of surface reflectances: assessments with hyperspectral images | For many fields of study and application, specifying surface reflectances in a simple and perceptually informative manner would be advantageous. We assessed whether a sensory reflectance matrix, which is a 3 x 3 matrix, could be used to approximate how a surface reflectance spectrum modulates the sensory colour signal across illuminants. While the reflectance spectrum is a physicist’s characterisation of a surface, telling us how the whole spectrum of incoming light will be affected by the surface, the sensory reflectance matrix is a sensory characterisation of the surface which tells us for any incoming light how the measures taken by the three cone photoreceptors will be affected by the surface. Using a 4-Alternative-Forced-Choice discrimination task, we tested whether observers could discriminate between the model’s approximate and accurate spectral renderings of hyperspectral images under narrowband and naturalistic, broadband illuminants for 8 hue directions. Discriminating the approximate from the spectral rendering was possible with narrowband, but almost never with broadband illuminants. These results suggest that our model specifies the sensory information of reflectances across naturalistic illuminants with high fidelity, and with lower computational cost than spectral rendering. We also found a strong correlation between deviations from cone excitation ratios and deviations from sensory reflectance matrices. It would be interesting to discuss the potential relationship of cone excitation ratios and sensory reflectance matrices as well as the relationship between sensory reflectance matrices and colour constancy. |
| Ilgin Cebioglu & Yesesvi Konakanchi | Exploring the visual and non-visual effects of light using spectrally-tuneable LEDs | Light is not only pivotal for visual perception, but also for essential functions such as sleep, nutrition and mood regulation. Understanding how light affects humans through visual and non-visual pathways has gained importance due to increased everyday exposure to artificial light from technological devices. Spectrally-tuneable LED light technology allows us to measure, quantify and manipulate light conditions to investigate visual and non-visual effects of light. For example, the colour appearance of objects may be altered while maintaining the visual appearance of the light, through controlled variations in the spectral composition of light. Using psychophysical methods combined with psychometric tests and controlled illumination variations, we investigate how our extensive experience with natural illumination shapes perceptual phenomena such as colour constancy and memory colour. Furthermore, we study whether illumination can be used to improve perceived palatability and nutritious quality of food items. |
| Lemona Xinxuan Zhang | Colour can be decoded from MEG frequency power | Colour is an indispensable quality of our visual experiences. While colour representation is relatively well understood in early visual pathways, we do not know as much about how different colours are encoded in the brain. Previous studies have investigated cortical colour representation by decoding colour from EEG or MEG time series data, which are limited by their relatively low signal-to-noise ratio and dependency on precise stimulus onsets. We have instead used frequency power spectra from an openly available MEG dataset obtained by Rosenthal et al. (2021, Current Biology, 31(3), 515-526). Our preliminary results show that support vector machine classifiers can be trained to decode colours presented on single trials with high accuracy from frequency power spectra. Further results from different classifier approaches have the potential to reveal spectral power markers for the perception of different colours, at least within individuals. This could have important implications for the neural basis of colour perception in terms of how subjective colour experiences might arise at cortical levels. Our results can also provide promising avenues for applications in, for example, brain-computer interfaces for participants with reduced mobility, or the induction of colour percepts without real stimuli. The work was funded by the ERC grant 949242 COLOURCODE to JB. |
| Margaret Thibodeau | What is thermal taste? | Thermal taste status (TTS) describes an individual difference in taste perception, determined by warming or cooling an individual’s tongue. Thermal tasters (TT) experience taste sensations in response to changing temperatures, whereas thermal non-tasters (TnT) do not. Interestingly, TT also rate the intensity of suprathreshold taste (e.g., sucrose - sweet), some trigeminal stimuli (e.g., alum sulphate - astringent) and beverages higher than TnT. Several other differences between TT and TnT have been proposed such as, differences in olfactory responsiveness, temperature responsiveness, emotional responsiveness, food involvement, and food disgust sensitivity. The mechanism(s) underlying thermally elicited sensations have yet to be fully elucidated. Two peripherally mediated mechanisms have been suggested for thermally elicited sweetness. TRPM5, a cation channel that is highly expressed in taste receptor cells, can be activated when warming at similar temperatures to those used in TTS screening. In addition, when the sweet taste receptor dimer TAS1R2/TAS1R3 is inhibited, TT that normally experience thermally elicited sweetness do not. However, during TTS screening 75% of TT report other taste sensations, such as sour, salty, bitter, and metallic, for which peripheral mechanism(s) have yet to be identified. Differences in the cortical activation patterns of TT and TnT (taste, oral somatosensory and reward regions) suggest that a central gain mechanism may also be involved in TTS. More research is needed to fully understand the phenotypic differences observed between TT and TnT, and the mechanism(s) that underlie these differences. To this end, a literature review of TTS will be presented and current gaps in knowledge will be highlighted. |
| Michaela Trescakova | Depth estimation in real-world scenes | Human depth perception is typically investigated using simple shapes defined by a small number of depth cues. Using the SYNS dataset of natural scenes, we examined the time-course of depth estimation, and the relative contribution of visual features including elevation, binocular disparity and colour to ordinal and ratio depth judgments. Participants viewed briefly presented images (17 – 267 msecs) drawn from 19 outdoor categories of SYNS images under monocular (Experiment 1) or both monocular and binocular (Experiment 2) conditions. Superimposed on the images were two crosshairs that identified a pair of locations in the scene. Participants determined which of these two locations was further away, and then used a slider to report the depth of the near location as a percentage of the depth to the far location. The depth difference and mean depth of the probed locations varied from trial to trial. In Experiments 3 and 4, we also manipulated the colour of the images (natural, unnatural and greyscale), transformed using CIELuv colour space. In all experiments, when elevation cues were informative, they dominated ordinal and ratio depth responses. Participants were able to estimate local depth even with extremely brief presentation durations and both elevation and binocular disparity informed these early depth estimates. Colour effects interacted with other depth cues such as binocular disparity, presentation time and elevation. Humans outperformed multiple computational models based on these cues, including KNN and semantic segmentation models, especially when the elevation cue was absent, and for longer presentation times. |
| Natasha Sigala | A case of secondary clinical lycanthropy in Huntington’s disease | We describe the case of a patient diagnosed with Huntington’s disease (HD), who, following a two-year history of anxiety with obsessional preoccupations, developed psychosis with clinical lycanthropy: a prominent delusional idea that he was a werewolf. Although there was no benefit from various antidepressants and antipsychotics, there was remarkable improvement of his symptoms following prescription of Clozapine. His choreiform movement disorder also improved as his mental state settled. Although some reported cases of clinical lycanthropy are related to neurological conditions, this is the first case in a patient with HD. We also discuss the relevance of cultural and personal factors in the expression of a delusion that incorporates disgust, and the potential role of somatosensory aberrations and misidentification of self. Some of his symptoms, such as his sense that he was shrinking, or his apparently perceiving a hypodermic needle as larger than his arm, support the idea that sensory misperceptions were a key element in the development of symptoms. We suggest that the idea of “secondary clinical lycanthropy” is relevant – the notion that lycanthropic delusions can arise from somatic hallucinations and/or alterations in the sense of physical identity in people with psychotic illnesses. Here we discuss the relevance of cultural and personal factors in the expression of a delusion that incorporates disgust, and the potential role of somatosensory aberrations and misidentification of self. |
| Nicholas Root | Synesthesia as a conscious experience of linguistic representation: a case study (diacritical marks) and call for international collaboration | When grapheme-color synesthetes look at linguistic symbols (such as the letters written on this page), they experience them as having a consistent, specific color. The colors that synesthetes experience are not random - they are influenced by numerous factors, many of them related to properties of the synesthete’s native language. Researchers have often pointed out that grapheme-color synesthesia is not merely a perceptual phenomenon, but also a psycholingusitic one: synesthetic associations are plausibly related to the computations and mental representations that underlie the reading process. The Cross-Language Synesthesia Consortium is a collaboration between synesthesia researchers around the world: we have collected data from synesthetes in more than a dozen different languages, with the dual aim of using language as a tool to study synesthesia, and using synesthesia as a tool to study language. Next year, we will release our dataset and full results to the public. Here, we present a small example of the type of question that can be asked of cross-language synesthesia data. We examine the role of diacritical marks in three languages around the world. First, we show that umlauts in German, which have a predictable influence on vowel formant (acoustic) structure, also have a predictable influence on synesthetic color. In Bengali, diacritical marks on consonants indicate the accompanying vowel, and the vowel’s synesthetic color ““pulls”” the color of the consonant towards it. In Arabic, diacritical marks also indicate which vowel accompanies a consonant, but these diacritical marks are usually omitted from text: can ““implicit”” linguistic symbols still influence synesthetic color? |
| Nicolas Rothen | Enhanced perception and memory: Insights from synesthesia and expertise | Memory research has identified many strategies to enhance memory. By contrast, natural foundations of enhanced memory remain to be explored. Numerous studies show that synesthesia is associated with enhanced memory. It has been hypothesized that wider changes in visual perception are closely linked with enhanced memory functions in synesthesia. However, the hypothesis has never been directly put to the test. In a first study, we investigated whether visual perceptual abilities in synesthesia are associated with better memory performance by comparing synesthetes who experience colors for letters with non-synesthetic color experts and non-synesthetic individuals from the more general population. Our results showed that color-synesthesia and color-expertise share a common profile of enhanced visual perceptual ability and memory in contrast to non-synesthetic individuals from the more general population. The findings suggest that visual perception and visual memory are more closely connected than previously thought. |
| Taysa-Ja Newman | The contribution of fractal dimension to infant and adult visual preference and adult aesthetics. | mage fractal complexity has been shown to contribute to visual aesthetics and preference. Adult preference for certain fractal dimensions in both natural and artistic images has been extensively investigated (Spehar et al., 2003; Spehar & Taylor, 2013). However, the role of fractals in infants’ visual preference is unexplored. Here we investigated infant visual preference for specific fractal dimensions as well as the relationship between infant visual preference and adult fractal aesthetics. The stimuli were black and white images of equal luminance generated using MatLab with fractal dimensions ranging from 1.1-1.9. Adults and 4-9 month old infants were shown 40 pairs of images and eye movements were recorded. Additionally, adult preference was measured using a paired forced-choice task. Both adult and infant mean looking time was significantly longer for the higher than lower fractal dimensions. However, adults’ forced-choice preference displayed a quadradic trend with highest preference for medium fractal dimensions (e.g. 1.5), in line with previous studies (Robles et al., 2020; Spehar et al., 2003). A k-means clustering analysis revealed three adult groups with distinct patterns of preference, one of which was the same as that found for infant and adult looking measures. We discuss the relationship between looking and liking, the implications for understanding how fractal dimension contributes to aesthetics, as well as the implications for visual development more broadly. |
| Tjark Andersen | The influence of viewing sensory-specific food pictures on physiology and eating behaviour: The FOODPIC study | Exposure to visual food content is implied in the development of obesity, at least in part due to increasing the viewers’ appetite. This increase in appetite appears to be sensory-specific to the viewed food, e.g., exposure to sweet food increases the desire for sweet food. However, recent research on mental imagery has demonstrated that repeated imagined consumption of food content can counteract this aforementioned increase in appetite. Such changes in appetite may be reflective of changes in the underlying physiology. However, the specific physiological research on food image viewing has been inconclusive, and sensory-specificity has not been studied in the context of food image viewing at all. The randomized cross-over study FOODPIC (ClinicalTrials.gov NCT05719142) is investigating how exposure to sensory-specific visual food content influences self-reported, i.e., interoceptive, appetite, its underlying physiology, and actual eating behaviour. Participants attend three visits in random order. During each visit, participants view either sweet food, savoury food, or non-food images for 15 minutes. Participants rate their (sensory-specific) appetite before and after the picture viewing. Furthermore, the participants have a total of nine blood samples drawn throughout the picture-viewing process to determine physiological changes in glucose and insulin concentrations, as well as other appetite hormones. Finally, as an eating behavioural endpoint, the participants select and consume a breakfast meal from an isocaloric menu comprised of two sweet and two savoury options across four portion sizes.Overall, the results of this study will shed light on the sensory-specific effects of visual food content exposure on interoception, physiology, and behaviour. |
| Veronica Pisu | Shape-related size biases in visual area judgements | Every day, we effortlessly interact with a wide range of objects of different size and shape, yet shape-related size biases have long been reported. For example, in area judgements, observers consistently judge triangles as larger than same-area squares and disks. Several authors have proposed that the biases are explained by the magnitude of one geometric features of the object, or the use of heuristic strategies (e.g., combining the object’s height and width), but the picture is complicated by inconsistent conclusions from studies focusing on specific subsets of shapes. Here we explored biases in area perception, across different shapes, allowing us to test the influence of a wide range of geometric features (height, compactness, elongation, etc.). In four online experiments, observers made 2AFC judgements (“which stimulus has larger area?”) for pairs of objects that differed in shape and / or area and / or orientation. We found clear shape-related biases: for example, triangular shapes were systematically perceived to have larger area than square shapes; disks were perceived to have the smallest area. When orientation was manipulated, the same shape was perceived to be larger or smaller depending on its orientation. Overall, no single geometric feature – or simple combination of two features ¬– provides a good correlate of the biases across all shapes, and a more complete description incorporating additional geometric features is needed to explain shape-related biases in area perception. We provide a multi-predictor model that quantitatively predicts biases in perceived area across 76 shape / size / orientation combinations. |